Informational Content of Trading Volume and Open Interest

– An Empirical Study of Stock Option Market In India

Sandeep Srivastava •

I. INTRODUCTION

Over the past three decades, option contract – defined as a contract that gives the holder (known as option
buyer) the right to buy or sell an underlying asset in future at a pre-agreed price – has been widely accepted as
one of the most useful derivative securities. This security was assumed to be redundant and valued with a no-
arbitrage relation in Black-Scholes model. The authors of this model argued that a portfolio comprising of the
stock and (riskless) bond would replicate the option position. However, several research studies, conducted
later on, have concluded that option market has been useful in improving the quality of underlying asset
market. More specifically, they have emphasized that this market plays an important role as a trading vehicle
which provides high liquidity, low trading costs, leverage, limited loss potential opportunities to option buyers
and least restrictions (like uptick’ rule for short sales). Most of these studies have argued in favour of robust
interaction between option and cash market that results in transmittal of information from one market to
another and thus is mutually beneficial resulting in better price discovery.
Empirical studies, more often, involve investigation of options’ effect on stock price behaviour at
two points in time separated by an event (like option listing and option expiration) that might affect this
behaviour. The volume of trading in option market is also found to be of lot of relevance in understanding
the price discovery function. It has been argued that on one hand price impounds information about the
average level of trader’s private information while on the other, volume captures signals relating to the quality
of trader’s information.
Extending the reasoning mentioned above, Easley, O’Hara and Srinivas (1998)1 investigate the
relationship between the option volume and the stock price changes to predict the informational content of
volume in option market. They arrived at two fold conclusions: (1) stock prices lead option volumes; and (2)
option volumes lead stock price changes. Their second finding is the basis for contention that option market
is an important venue for information based trading. Recently, Bhuyan and Chaudhury (2001)2 have
examined the role of option market’s open interest in conveying information about the future movement of

•
Assistant Professor, Finance and Accounting, Lal Bahadur Shastri Institute of Management. The views expressed and
the approach suggested in this article are purely personal and not necessarily of his employer or NSE
1For details, see Easley, David, Maureen O’Hara and P.S. Srinivas, 1998, Option volume and stock prices: Evidence on
where informed traders trade, Journal of Finance, 53, 431-465.
2For details, see Bhuyan, Rafiqul and Mo Chaudhury, 2001, Trading on the information content of open interest:
Evidence from the US equity options market, Working Paper, McGill University.
the underlying asset and have shown that the trading strategies based on this predictor fetches superior results
as compared to traditional strategies. Adding on to this, Bhuyan and Yan (2002)3 develop several price
predictors from the open interest and volume of trading at different strike prices, and conclude that they
exhibit significant explanatory and predictive power for the future stock prices.
This paper has its base in their results and is among one of the earlier attempts to study the inter-linkages
between the option market and cash market. It seeks to examine the role of option market’s open interest and
volume of trading in determining the underlying share prices in Indian stock market. This study is significant
because Indian market is prone to manipulations and raiding by operators time and again. Through this study,
the author seeks to provide a valuable input to the uninformed investors using which they can discover the
price of underlying asset in a more efficient as well as effective manner. The results are encouraging and
provide a basis for further examination of different variables in option market and their interrelationships
with the underlying stock prices. The remainder of this paper is divided into three sections. Section II
contains the data and methodology used in the study. Section III elaborates the empirical results followed by
the last section that includes the conclusion and direction for future research.

II. DATA AND METHODOLOGY

Securities and Exchange Board of India (SEBI) introduced stock options on 31 shares in July 2001. The
trading interest in these contracts has been consistently increasing since then as shown in figure 1. The option
contracts on ICICI Ltd. and Reliance Petrochemicals Ltd. ceased to exist after their merger with ICICI Bank
and Reliance Industries Ltd. respectively that left the total number of optioned stocks to 29. Not all the
option contracts are liquid at National Stock Exchange (NSE). Therefore, our sample consists of options on
15 individual stocks that are most liquid based on the trading volume. Table 1 shows the list of stocks
included in this study alongwith their lot sizes. Later on, SEBI expanded the list to include 20 more stocks
but that has been excluded in this study, as the data on these stock options is not available for the entire
period of study.
The data for this study was taken from the daily bhavcopy posted on the NSE website. It provides all
the market information on call and put options traded on different stocks during the day that include option
premium (open, high, low and close), trading volume and open interest at each strike price. This study covers
stock option contracts for four months, namely November 2002, December 2002, January 2003 and February
2003, comprising a total of 77 trading days (excluding the expiration day).
Currently, we have stock option contracts available in Indian market for one-, two- and three- month
maturity. As the present study attempts to decipher the price-volume relationship, the liquidity of stock
options becomes an important issue. Keeping this in view, we have included one-month contracts only

3For details, see Bhuyan, Rafiqul and Yuxing Yan, 2002, Informational role of open interests and volumes: Evidence from option markets, Paper
presented at Twelfth Annual Asia-Pacific Futures Research Symposium held in Bangkok on December 3-4, 2001.
because they are most liquid. Further, the expiration day data has been excluded from the study to avoid the
biasness due to expiration effect.
As mentioned earlier, this study investigates the significance of net open interest and trading volume in
stock option market to predict the underlying stock prices. The methodology used here has been taken from
Bhuyan and Chaudhury (2001) and Bhuyan and Yan (2002). The notations used are the same as have been
used in these studies. The call option open interest based predictor (COP), put option open interest based
predictor (POP) and volume based predictors, both for call and put option, have been calculated by
multiplying the strike price with the weights of open interest or trading volume for the respective strike
prices. Thereafter, the relative significance of each of these predictors has been examined using a linear
regression model having future spot price as endogenous variable and current spot price, time to expiry, and
each of the open interest and volume based predictor as exogenous variables. The natural logarithms of
variables have been used to rescale data to ‘pull in’ extreme observations and handle heteroscedasticity. The
relative significance of open interest and volume based predictors has also been examined separately.
III. EMPIRICAL RESULTS
This study includes the data for 15 most liquid options on individual shares traded at NSE and the regression
analysis is carried out for all the 15 shares in a consolidated form as well as separately. This has been done to
check if there is a significant variation in the results while taking each stock option independently.
An examination of the results of regression analysis reveals that both the open interest and volume
based predictors are significant explanatory variables for estimating the future spot price of underlying shares.
However, as against the findings of earlier study, Bhuyan and Yan (2002), the coefficients of call as well as put
option open interest based predictors are negative and statistically significant. On the other hand, the
coefficients of volume based predictors for call and put options are positive and statistically significant.
Bhuyan and Yan (2002) find the positive coefficient for COP and negative coefficient for POP. They argue
that when investors expect the price of underlying stock to increase they would be willing to buy call options
at higher strike price and in case of anticipated decline in stock prices they would prefer to buy put options.
However, our findings contradict their conclusions. Still, the coefficient of POP is highly negative and more
statistically significant as compared to the coefficient of COP. The model could explain around 27 per cent of
the variations in future spot price of underlying shares. The findings have been identical even when the data
for the last five days upto expiration was excluded from the analysis.
To examine the informational content of trading volume and open interest in more detail, the study
has been extended to include the regression analysis of 15 individual stock options separately. These results
are quite insightful and out of the total 15 stock options covered in this study, 8 are found to have positive
coefficients for COP and negative coefficients for POP. They are statistically significant for both COP and
POP except for Satyam Computer Services Ltd. Also, the model has considerable explanatory power in these
cases. On the other hand, 6 stock options have negative coefficients for COP and positive coefficients for
POP. However, the results are not statistically significant except for Mahanagar Telephone Nigam Ltd. and
Tata Iron and Steel Company Ltd. One stock option, on Digital Globalsoft Ltd., has negative coefficient both
for COP and POP but it is not statistically significant.
These findings provide support in favour of the argument that the coefficient should be positive for
COP and negative for POP. This is because of the fact that informed investors would buy out-of-the-money
call options (at higher strike price level) in anticipation of rise in stock prices leading to increase in COP.
Similarly, they would prefer to buy put options at higher strike price when they have specific information
about the decline in stock prices, which would lead to higher POP. Another possibility may be that the
informed investors may like to write put options when they expect the stock prices to increase and call
options when they expect stock prices to decline.
It may be argued that the model used in this study is based on the basic premise that informed
investors are expected to go long only either on call options or put options. However, it is to be noted that I
have not ruled out the option writing by the informed investors but I believe that such transactions are going
to be less. The reason is that while writing options one would take exposure to the unlimited risk potential.
When informed investors are acting on the basis of information that would materialize in future, they are
already undertaking the calculated risk. They would avoid multiplying their risk exposure by going short in the
option market. Further, given the high degree of volatility in Indian stock market, the probability of informed
investors buying options rather than writing options would be more. Therefore, the COP is expected to have
positive coefficient while the same is expected to be negative for POP. It should be noted that the study
emphasizes upon the COP and POP and not the net open interest alone and hence it needs to be seen in the
light of both net open interest and strike prices. There is a likelihood that the open interest based predictors
may increase even when the net open interest declines. This is due to the fact that investors may be dealing in
the call or put options at higher strike price that leads to increase in these predictors. The reverse would
happen in case of increase in net open interest but the options are entered at lower strike price.
To draw the comparative analysis of predictive ability of call and put options open interest based
predictors and volume based predictors, the regression analysis is conducted for them separately The
explanatory variables, namely call and put option open interest based predictors and call option volume based
predictors, are found to be having negative coefficients. These are statistically significant only for the call
option predictors. The put option volume based predictor has positive coefficient but is not statistically
significant. These results do not offer much explanation of the price discovery behaviour except for the fact
that net open interest based predictors serve as better parameters for predicting future spot price of
underlying asset in comparison to the volume based predictors, which has been considered in the previous
studies including Easley, O’Hara and Srinivas (1998).
IV. CONCLUSION AND DIRECTION FOR FUTURE RESEARCH
This study has strengthened the argument forwarded by Bhuyan and Yan (2002) that net open interest of
stock option is one of the significant variables in determination of the future spot price of underlying share.
The results clearly indicate that open interest based predictors are statistically more significant than volume
based predictors in Indian context too. However, there are certain differences in the outcome of this study
and Bhuyan and Yan (2002). According to their study, the coefficient of COP has positive sign and the
coefficient of POP possesses negative sign. Based on these findings, they argue that the informed investors
would buy out-of-the-money call options when they expect the market to rise and put options when they
expect it to decline. The reasoning, as mentioned in the previous section, does support their contention.
However, I have given the additional explanation for this to happen which would be more relevant in context
of volatile markets like India. This explanation is with regard to the trading behaviour of informed investors
in Indian market.
In this study, both the coefficients of COP and POP are found to have the negative sign. This
finding is not as per the expectations. However, when I investigated the same for the 15 individual stock
options separately the evidence in favour of the above-mentioned argument is clearly visible. The coefficient
for COP is found to be positive while the same for POP is negative for majority of stock options and it is
statistically significant. Therefore, the results of this study show that option market, more specifically the net
open interest, is likely to be informative about the future movement of stock prices. Investors who do not
possess the specific information about the future price movement can use these predictors for deciding upon
their trading strategies. Even the sophisticated investors can explore these estimates and further refine their
trading strategies with more informational inputs.
There may be a number of reasons for the difference in results arrived in US context and India.
Firstly, the exchange traded stock derivative market in India is of recent origin and it takes time for the
investors to realize the true potential of these instruments. Secondly, the participation of institutional
investors in Indian stock derivative market is extremely limited. It can be attributed to the regulatory
restrictions wherein such investors are allowed to use derivative securities for hedging purposes only. It can
be proved from the fact that cumulative foreign institutional investor’s position as percentage of total gross
market position in the derivative segment was 12.39 per cent as on August 5, 2003. The story of mutual funds
is not significantly different from this. Therefore, the investors who have better access to information and can
be classified in the category of informed investors are constrained to deal in derivative securities. Though
there are some positive developments taking place in this direction, these securities are yet to gain significance
in the portfolio of institutional investors in India. Lastly, there are some issues relating to the accounting and
tax treatment of profit arising from dealing in derivative securities that further prohibit the wider participation
of such investors in derivative segment.
 Our research contributes to the existing literature by providing some evidence of option market
activity and the impact of non-price variables on the future spot price of underlying asset in context of a
developing nation like India where derivatives have been recently introduced in stock market. As the market
activity progresses further, I hope that more evidence would come in support of our findings. It would not
only benefit the investor community but also provides support for the hypothesis that derivative securities
enhance the quality of underlying asset market. This study also complements the earlier evidence documented
in Srivastava, Yadav and Jain (2002)4 with regard to the efficiency of stock index futures market in India. It
would definitely help the regulatory bodies in policy-making and further strengthening the efforts to promote
the derivative market in India. There are many areas which are still unexplored and can be addressed by the
future studies.
First and the foremost is the use of intraday data in determining the significance of net open interest
because many profitable opportunities are utilized within a course of day and hence future studies may take
up this aspect into consideration. I have used a sample of 15 near-month stock options for the present study
because they are most liquid. As the number of shares in stock option segment is increased to 49, the future
studies may be conducted with larger sample size and longer duration data. Lastly, the significance of daily
change in open interest in predicting the future spot price may be another area that can be explored by the
future studies.

4 For details, see Srivastava, Sandeep, Surendra S. Yadav and P.K. Jain, 2002, Early efficiency signals from stock index futures market in India, Paper

presented at 15th Australasian Finance and Banking Conference held in Sydney on December 16 – 18, 2002.
 Table 1: Sample Firms and Lot Sizes in Stock Option Market

No. Underlying Stock Contract Size (No. of shares)

1. Associated Cement Company Ltd. 1500
2. Bharat Petroleum Corporation Ltd. 1100
3. Digital Globalsoft Ltd. 400
4. Hindustan Lever Ltd. 1000
5. Hindustan Petroleum Corporation Ltd. 1300
6. Infosys Technologies Ltd. 100
7. Larsen & Toubro Ltd. 1000
8. Mahindra & Mahindra Ltd. 2500
9. Mahanagar Telephone Nigam Ltd. 1600
10. Reliance Industries Ltd. 600
11. Satyam Computer Services Ltd. 1200
12. State Bank of India 1000
13. Sterlite Optical Technologies Ltd. 600
14. Tata Engineering and Locomotive Co. Ltd. 3300
15. Tata Iron and Steel Company Ltd. 1800