Option listing, trading activity and the informational efficiency of the underlying stocks

Transcription

1 Option listing, trading activity and the informational efficiency of the underlying stocks Khelifa Mazouz, Shuxing Yin and Sam Agyei-Amponah Abstract This paper examines the impact of option listing on the informational efficiency of the underlying stocks. We show that option listing improves the speed at which information is incorporated into the stock price and reduces the pricing inefficiency of the underlying stocks. We attribute the efficiency gain to the arrival of better informed traders to the stock market. We also show the informational efficiency gain from option listing is negatively associated with the pre-existing market quality of the underlying stocks. This result reflects the fact that there is more potential to uncover private information in otherwise less efficient stocks. Our analysis also indicates that stock traders are better informed than option traders. This evidence is consistent with the migration of uninformed traders (i.e. uninformed hedgers and/or speculators with ill-founded but strong beliefs) from the stock to the option market. Keywords: Option listing, informational efficiency, informed trading, stock volume, option volume 1

2 Option listing, trading activity and the informational efficiency of the underlying stocks 1. Introduction A common assumption in option valuation models (including Black and Scholes, 1973) is that stock options are redundant securities as their payoffs can be replicated in continuous time by combining a position in a stock with risk-free lending and borrowing. This assumption implies that option listing should have no effect on the market quality of the underlying stocks. However, in incomplete markets, with transaction costs and trading constraints, stock option payoffs may not be easily replicated by dynamic trading strategies in simple securities such as stocks and bonds (Naik and Lee, 1990; Pan and Liu, 2003). The view that the stock option is a redundant security also ignores the informational role of the option. As Grossman (1988) argues, prices of real securities convey important information to market participants that will not be conveyed if the real security is replaced by synthetic trading strategies. Ross (1976) and Hakansson (1982) also argue that options improve the efficiency of incomplete asset markets by expanding the contingencies covered by the marketed securities. Cao (1999) also claims that the presence of options should improve stock market efficiency by allowing informed agents to trade more effectively on their information. Furthermore, informed agents may prefer to trade options rather than stock because of the inherent leverage associated with options and the ability of options to overcome the short selling constraints imposed on certain stocks (John et al., 1993). In contrast to the preceding arguments, options may attract agents with ill-founded but strong beliefs (Roll et al. 2010). These agents may also use options to overcome their borrowing and/or short selling constraints. If the information released by these agents has a negative influence on the ability of stock market participants to make inferences from prices (Stein, 1987), options would harm the informational efficiency of the underlying stocks. Given the contradiction between these opposing arguments, many researchers choose to investigate the interactions between options and stocks empirically. However, even the empirical research in the area of option listing has not yet reached a consensus. DeTemple and Jorion (1990), Fedenia and Grammatikos (1992), Kumar et al. (1998), Sahlstrom (2001), Faff and Hiller (2005) and Liu (2009), among many others, show that option listing increases 2

3 the price and trading volume but decreases the volatility of returns and bid-ask spreads of the underlying stock. However, Sorescu (2000) shows that options listed after 1980 are associated with negative price effects. Other studies, such as Harris (1989), Bollen (1998), Mazouz (2004) and Mayhew and Mihov (2004), use a control sample methodology to show that the volatility change associated with the optioned stocks is not significantly different from a matched set of non-optioned stocks after option listing. While a large body of the literature focuses on the price, volatility and liquidity effects of option listing, the impact of option listing on the informational efficiency has received comparatively less attention. Cao and Wei (2010) show that information asymmetry is greater for options than stocks, indicating that options are more appealing for informed agents. Easley et al. (1998) and Ni et al. (2008) show that options order flows contain information about both stock price and stock return volatility. Jennings and Starks (1986) and Mendenhall and Fehrs (1999) show that option listing increase the speed at which stock prices adjust to earnings announcements. Consistent with the notion that options improve the informational efficiency of the underlying stocks, Chern et al. (2008) find that optioned stocks exhibit a significantly lower price reaction to stock split announcement than non-optioned stocks. Kumar et al. (1998) also find that option listing improves the market quality of the underlying stocks. However, their study focus on a different set of measures including the variance of prediction errors, the weight placed by the specialist on public information in revising prices and the adverse selection component of the bid-ask spread. This study contributes to the literature on the impact of option listing on the informational efficiency of the underlying stocks by providing a different perspective on the issue. First, we adopt entirely different informational efficiency measures. 1 We initially measure informational efficiency using the partial adjustment parameter of Amihud and Mendelson s (1987) model. This parameter is a useful indicator of the speed at which information is incorporated into the stock price. Then, we estimate a pricing inefficiency parameter of Chelley-Steeley (2008) to determine the extent to which transaction prices deviate from intrinsic values. Second, we relate the potential impact of option listing on the informational efficiency of the underlying stocks to the level of trading activity in options. Roll et al. (2009) argue that the valuation benefit from options to a firm should depend on option trading activity, over and beyond the presence of an options market on the firm s stock This is because traders, both informed and uninformed, may not be attracted to an 1 These measures are used previously by Chelley-Steeley (2008) and Chelley-Steeley and Skvortsov (2010) in the context of market microstructure. 3

4 inactive option market. Thus, the effect of option listing on the informational efficiency should depend on the level of options trading activity. Third, we test whether the informational efficiency impact of option listing depends on the trading activity in stocks. Chelley-Steeley (2008), for example, shows that changes in market microstructure have stronger impact on illiquid stocks. Finally, we examine whether the pre-existing informational efficiency influences our findings. Easley et al. (2002) show that the impact of options trading is greater in stocks with low analyst following and high values of the probability of informed (PIN) trading measure. Their evidence indicates that the impact of option trading on information production depends on the level of information asymmetry in the pre- listing period. Our results suggest that after option listing stock prices adjust more quickly to new information and pricing inefficiency of the underlying stocks declines. We attribute the efficiency gain to the possibility that option listing attracts better informed traders to the stock market. The cross-sectional regression analysis reveals a negative association between the pre-existing market quality and the change, across pre- and post-listing periods, in the informational efficiency parameters. This result supports Easley et al. s (2002) argument that more private information could be uncovered in less efficiently priced stocks. The regression results also indicate that stock traders are better informed than option traders. We attribute this finding to the migration of uninformed hedgers and/or speculators with ill-founded but strong beliefs from the stock to the option market. The remainder of the paper is organized as follows. Section 2 describes the methodology. Section 3 describes the data and provides some descriptive statistics. Section 4 presents our results and Section 5 concludes. 2. Methodology This section presents the methods used to estimate the partial adjustment parameter and the pricing inefficiency index The partial adjustment parameter Amihud and Mendelson (1987) develop a price formation model that distinguishes between the intrinsic value of a security i at time t, V it, and its observed price, P it., (1) 4

5 , (2) where p it and v it are the logarithmic values of P it and V it, respectively; g i is the partial adjustment coefficient of a security i with a value 0 < g i < 2; is a normally distributed error term with a zero mean and constant variance ; m i is the expected daily value of return and is an independent and identically distributed (i.i.d) random disturbance with mean of zero and variance. The magnitude of the parameter g i indicates the adjustment of transaction prices towards their intrinsic values. Specifically, g i = 0 implies that the observed price does not react to changes in the intrinsic value, 0 < g i < 1 suggests the observed price adjusts partially to changes in the intrinsic value (underreaction to the arrival of new information), g i = 1 indicates that the current transaction price adjust fully to the changes in the asset s intrinsic value and g i > 1 represents traders overreaction to the arrival of news. Equation (1) can also be expressed as (3) If we substitute with and with, equation (3) becomes, (4) Since follows a random walk with drift,, or, also follows a random walk with drift. Thus,, (5) where the drift term is equal to and. Since the variance of is, the variance of is. Following Chelley-Steeley (2008) and Chelley-Steeley and Skvortsov (2010), we use Kalman filter and smoothing algorithms of Harvey (1993) to estimate the parameters of equations (4) and (5). After obtaining and from equations (4) 5

6 and (5), we can calculate the price adjustment parameter,, and the asset s intrinsic value,. Equations (4) and (5) are estimated for each stock i over the [-261, -1] and [0, 260] windows around option listing. If options improve the informational efficiency of the underlying stocks, the parameter will move towards unity after option listing Pricing inefficiency index Another informational efficiency measure used in this study is based on the extent to which the observed transaction prices deviate from their intrinsic values. In line with Chelley- Steeley (2008) and Chelley-Steeley and Skvortsov (2010), we define the pricing inefficiency (or pricing error) of a security i at time t as (6) and the average pricing inefficiency of a security i over a period from t =1 to T is given as (7) The PI i statistic is also estimated for each stock i over the [-261, -1] and [0, 260] windows around option listing. A lower value of PI i after option listing will indicate that options improve informational efficiency of the underlying stocks as they bring transaction prices closer to their intrinsic values. 3. Data and sample description The list of the dates on which option contracts were listed on individual stocks, between January 1978 and December 2007, is obtained from the London International Futures and Options Exchange (LIFFE) website. The list includes a total of 186 stocks. To be included in the analysis, the optioned stock must have a complete set of 260 daily price observations available from DataStream on both side of the option listing date. Several stocks were excluded due to insufficient observations. The final sample contains 105 optioned stocks. Table 1 presents the number of new options included in our sample each year between 1978 and

7 A control sample approach is adopted to isolate the effect of factors other than option listing. We match each optioned stock with a non-optioned stock with the same size decile, but not necessarily the same industry, at the time of the option listing. 2 To be considered as a control stock, the non-optioned stock must not experience option listing within the next 260 days. Data on daily turnover by volume, market capitalisation, book-to-market value and option volume is obtained from DataStream. Panel A of Table 2 reports a summary of the pre-listing characteristics of the sample of optioned stocks and the control sample. Except for the variance of returns, our statistical tests confirm that the optioned stocks and the control stocks have similar characteristics before the introduction of options. 3 Panel B of Table 4 also provides some descriptive statistics of the option volume measures. Until February 1991, option volume data was not available from DataStream. Since almost half of the option listing events occurred prior to 1991, only 45 of our sample stocks have option volume data available immediately after option listing. [Insert Tables 1 and 2 about here] 4. Empirical results This section reports our findings on the impact of option listing on the speed of price adjustment to new information and the pricing inefficiency of the underlying stocks. It then presents the results from the cross-sectional regression analysis, which relates the efficiency change to the characteristics of the underlying stocks and the trading activity in options and stocks Changes in the informational efficiency parameters Table 3 presents the pre- and post-option listing estimates of partial adjustment parameter (g i ) and pricing inefficiency measure (PI i ) for both optioned stocks and their matched pairs. Panel A of Table 3 focuses on the parameter g i. The pre-listing partial adjustment parameters associated with the sample of optioned stocks and the control sample 2 We use both industry and size in the selection of control stocks whenever possible. However, we choose to sacrifice industry effect for the size effect, when a suitable control stock from the same industry and size decile is not available. 3 Volatility differences between the main and the control sample is driven by outliers. These differences disappear completely after excluding the 5 most volatile optioned stocks and their matched pairs from the analysis. 7

8 average to and , respectively. The average values of g i are below unity, indicating that stock prices underreact to the arrival of news in the pre-listing period. In the post-option listing period, the cross-sectional averages of the estimated partial adjustment coefficients for the sample of optioned stock and the control sample are and 0.773, respectively. About 60% of the optioned stocks and 57% of the control stocks experience rise in g i after option listing 4. Thus, a significant proportion of stocks in the optioned sample exhibit an increase in the speed of price adjustment after option listing. Both t-test and Wilcoxon Signed Rank test also indicate that the change in g is significant for the sample of optioned stocks, but not for the control sample. As the matched sample of nonoptioned stocks fails to show any significant increase in the speed of price adjustment, we attribute the significant increase in the optioned stocks to the event of stock option listings. Thus, option listing improves the informational efficiency by causing the price adjustment parameter to move towards unity. [Insert Table 3 about here] Panel B of Table 3 presents the pre- and post-listing estimates of the pricing inefficiency index (PI i ). The pre-listing mean PI i values of the optioned and control stocks are and 0.390, respectively. In the post-listing period, optioned stocks experience a statistically significant average decline of in the mean PI i. This decline is unlikely to be driven by extreme values, as 63% of the optioned stocks exhibit PI i decrease following option listings. The matched control stocks do not experience any significant change in their pricing inefficiency. Both paired t-test and Mann-Whitney test indicate that the PI i decline in the sample of optioned stocks remains significant after accounting for the corresponding PI i change in the control sample. This result implies that the change in the pricing inefficiency of the optioned stocks is not driven by factors other than option listings. Consistent with our results on the impact of option listing on the partial adjustment parameter, option listing also improves the informational efficiency by brining the observed stock prices closer to their intrinsic values. This evidence remains valid even after accounting for the corresponding changes in the PI i of the control stocks. 4 We test the significance of these proportions using a simple binomial test of proportionality. The corresponding test statistic is. Where p is the proportion of observations for which the post-listing estimate of g is greater than its pre-listing estimate. 8

9 [Insert Table 4 about here] Table 4 shows that optioned stocks experience a significant volume increase of 19% after option listing. The volume increase associated with the optioned stocks is not driven by the market-wide conditions, particularly since the volume of the control stocks does not exhibit any significant change after option listing. Kumar et al. (1998), Sahlstrom (2001) and Faff and Hiller (2005) also show that option listing is associated with a significant stock volume increase in the US, Finland and UK, respectively. The significant increase in the stock volume indicates that option listing attracts new traders to both the stock and the option market. Indeed, the presence of option contracts as efficient hedging and speculative vehicles is likely to attract new traders, who may find the option market more cost effective than the synthetic option trading strategies. Our results are largely consistent with the previous literature (e.g., Jennings and Starks, 1986; Kumar et al., 1998; Mendenhall and Fehrs, 1999; Sahlstrom, 2001), which shows that options increase the amount of private information conveyed by stock prices. In what follows, we examine whether the gain in the informational efficiency is related to the firm characteristics and the trading activity in the stock and the option market Cross-sectional regression analysis To examine the determinants of the change in the informational efficiency, we regress change in each informational efficiency measure ( or ) against a set of explanatory variables, the level of informational efficiency in the pre-listing period ( or ), the percentage change in trading volume ( ) and post-listing option volume ( ). 5 We explain and provide some justifications for each of the independent variables below. and are stock i s pre-listing price adjustment coefficient and pre-listing pricing inefficiency index, respectively. These variables are included in the analysis to examine whether the option listing impact on the informational efficiency depends on the pre-existing market quality of the underlying stocks. Since there is more potential to uncover private information in less efficiently priced stocks, we expect a negative relationship 5 We also include average pre-listing market capitalization ( ), the average pre-listing daily stock volume (VOL i,pre ) and a time dummy ( ) in the cross-sectional regression. The results are not reported, as none of the three variables is statistically significant. More details are available upon request. 9

10 between the pre-existing market quality and the change, across pre- and post-option listing periods, in the informational efficiency parameters. is the percentage change in trading volume measured as, where and are the stock i s pre-listing and post-listing average daily volume, respectively. This variable is also expected to affect the efficiency of the underlying stocks. If the existence of options attracts more traders to the stock market, the overall effect of stock volume increase on the informational efficiency of the underlying stocks would depend on quality of information possessed by newly attracted traders. The variable is the post-listing average daily volume of the option contract listed on a stock i. We use post-listing average daily put option volume (VPM i,post ), post-listing average daily call option volume (VCM i,post ) and the post-listing average daily aggregate option volume (VM i,post ) to reflect the trading activity in the option market. Roll et al. (2010) argue that active options markets stimulate greater informational efficiency by encouraging more informed trading. However, options would harm stock pricing efficiency if options trading activity simply represent uninformed speculation (Roll et al., 2009). Given the theoretical ambiguity, the relationship between and informational efficiency remains an unsolved empirical issue. To account for the potential effect of serially correlated residuals of the crosssectional regressions, all the regressions are estimated using Newey-West serial correlation and heteroskedasticity adjustments. Table 5 reports the cross-sectional regression results of the change in the informational efficiency measures regressed on the selected determinant variables. Panel A of Table 5 reports the results for as the dependent variable. Models 1 through 3 show that both and are negatively correlated with. The significantly negative coefficient on implies that stocks with low pre-listing price adjustment coefficients benefit more from the introduction of options. This finding is also consistent with the view that the impact of option trading on information production is larger for less efficiently priced stocks (Easley et al. 2002; Roll et al. 2009). However, the negative 10

11 association between option volume and stock efficiency indicates that option trading activity is dominated with uninformed traders 6. The coefficient on % is positive and statistically significant in Models 4 through 6. This finding suggests that the speed of price adjustment to information increases with stock market liquidity. Models 4 through 6 also indicate that, in the presence of %, the significance of weakness. This evidence suggests that the gain in price efficiency comes from trading in stocks rather than options. In other words, option listing is likely to contribute to the stock price efficiency by stimulating information production in the stock market. [Insert Table 5 about here] Panel B of Table 5 presents the results of regressed on different sets of explanatory variables. The coefficient on is negative and significant across all the models, indicating that less efficiency priced stocks benefit more from the price noise reduction associated with the introduction of options. The results of OLS estimation of Models 1 through 3 suggest that is significantly related to VPM i,post and VM i,post, but not related to VCM i,post. These results imply that put option traders increase the amount of noise in the stock price. The significantly negative association between and %, observed in Models 4 through 6, indicates that the increase in stock volume brings transaction prices closer to their intrinsic values. The statistical significance of the option volume variables disappears completely in the presence of % (see Models 10 through 12), suggesting that the efficiency gain from trading stocks outweighs the noise induced by the option market traders. In short, the cross-sectional regression analysis suggests that stock traders possess better information than option traders. These results can be attributed to the migration of stock traders to the option market. Stock option contracts are regarded as efficient risk management vehicles contributing to the effectiveness of investment allocation (Cox and Rubinstein, 1985). Stock traders are no longer obliged to sell their holdings to protect themselves against expected inverse stock price fluctuations. They can simply buy a put option contract to hedge against this possibility. The inherent leverage associated with 6 See below for a detailed discussion on this point. 11

12 options and the presence of short-sell restrictions on certain stocks may also lead to the migration of speculators from the stock to the option market. Since stock traders possess better information than option traders, it is possible to argue that traders who migrate to the option market are less informed than those who choose to remain in the stock market. In other words, our findings suggest that uninformed hedgers and speculators with ill-founded but strong beliefs may find trading options more attractive than trading stock. 5. Conclusion This paper examines the impact of option listing on the informational efficiency of the underlying stocks. Amihud and Mendelson s (1987) partial adjustment parameter model is used to capture the speed at which information is incorporated into stock prices and Chelley- Steeley s (2008) pricing inefficiency index is employed to measure the extent to which the observed stock prices deviate from their intrinsic values. We use a control sample methodology to control for the informational efficiency changes caused by factors other than option listing. We show that option listing enhances the speed of stock price adjustment to new information and reduces the pricing inefficiency of the underlying stocks. Our results are not driven by changes in the market-wide conditions, as the informational efficiency parameters of the control sample remain unchanged after option listing. We attribute the observed increase in the informational efficiency to the arrival of better informed traders, who may have found option contracts more cost effective than the synthetic option trading strategies, to the stock and the option market. These results are supportive to the findings of number of existing studies, such as Jennings and Starks (1986), Kumar et al. (1998), Mendenhall and Fehrs (1999) and Sahlstrom (2001), which suggest that option listing increases the amount of private information conveyed by stock prices. The cross-sectional regression analysis suggests that the market quality of less efficiently priced stocks benefits more from the introduction of option. This result is expected since there is more scope to uncover private information in less efficiently priced stocks. We also show that stock traders possess better information than option traders after option listing. We attribute this discrepancy in the information quality to the migration of uninformed hedgers and/or speculator with ill-founded but strong beliefs from the stock to the option market. 12

13 Table 1 The yearly distribution of the sample of optioned stocks Year Number of new listings in our sample Year Number of new listings in our sample

14 Table 2 Summary statistics of the sample of optioned stocks and the control sample Panel A: The pre-listing characteristics of optioned and control stocks Return i,pre Variance i,pre VOL i,pre SIZE i,pre B/M i,pre Mean Median Mean Median Mean Median Mean Median Mean Median Optioned stocks (N = 105) Control stocks (N = 105) Optioned vs. control p-value for t-test ** P-value from Mann-Whitney test ** * Panel B: Descriptive statistics for the option volume data VCM,I,post VPM i,post VM i,posy N Mean Median Low High , Q Q Q Q Notes VOL i,pre is turnover by volume (in thousands of pounds), SIZE i,pre is the market capitalization (in millions of pounds), B/M i,pre is the book to market ratio, VCM i,post is the call option volume, VPM i,post is the put option volume, VM i,post is the aggregate option volume (VM i,post = VCM i,post + VPM i,post ). The stock s average return, variance, VOL i,pre, SIZE i,pre and B/M i,pre is measured over the [-261, -1] window whereas VCM i,post, VPM i,post and VM i,post are measured over the [0, 260] window around the option listing. Q1, Q2, Q3 and Q4 are the average option volume of the first, second, third and forth quartiles, respectively. ** and * indicate significance at 5% and 10%, respectively. 14

15 Table 3 Changes, between pre- and post-option listing periods, in the partial adjustment and pricing efficiency parameters Panel A: Partial adjustment parameter (g i ) around option listing Main Control Main vs. Control N Mean (median) pre (0.837) (0.818) Mean (median) post (0.859) (0.823) Mean (median) change ** (0.049 ** ) (0.015) ** (0.015 ** ) t-statistic (non-parametric 1 ) (2.348) (0.027) (2.257) % increase 2 60% ** 57% Panel B: Pricing inefficiency index (PI i ) around option listing Main Control Main vs. Control N Mean (median) pre (0.311) (0.241) Mean (median) post (0.221) (0.208) Mean (median) change *** ( *** ) (-0.029) *** ( ** ) t-statistic (non-parametric) (-2.876) (-0.730) (-2.326) % decrease 63% *** 54% Notes 1 We use Wilcoxon Singed Rank test to examine for the change within the same sample and Mann-Whitney test to compare the changes across the two samples. 2 % increase in Panel A refers to the percentage of stocks experience increase in g i after option listing and % decrease in panel B is the percentage of stocks exhibiting a decline in PI i after option listing. We test the significance of these proportions using a simple binomial test of proportionality. The corresponding test statistic is. Where p is the proportion of observations for which the post-listing estimate of g i (PI i ) is greater (less) than its pre-listing estimate. ***, ** and * indicate significance at 1%, 5% and 10% levels, respectively. 15

16 Table 4 Trading volume around option listing Main Control Main vs. Control Mean (median) Pre 5, (2,727.46) 5, (1,580.66) Mean (median) Post 6, (3,596.13) 5, (1,571.04) Mean (median) Change * ( *** ) (-9.62) 1, *** ( *** ) t-statistic (non-parametric 1 ) (3.216) (-0.345) (2.870) Notes 1 We use Wilcoxon Singed Rank test to examine the statistical significance of the change in the informational efficiency associated with each sample and Mann-Whitney test to compare the average change in the informational efficiency measures across the two samples. ***, ** and * indicate significance at 1%, 5% and 10% levels, respectively. The level of significance is based on the values from the non-parametric tests. 16

17 Table 5 Cross-sectional regression results Panel A: g i as the dependent variable Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Coef (t-stat) Coef (t-stat) Coef (t-stat) Coef (t-stat) Coef (t-stat) Coef (t-stat) Intercept *** (6.210) *** (7.360) *** (6.800) *** (4.270) *** (5.010) *** (4.450) g i,pre *** (-6.560) *** (-7.710) *** (-7.250) *** (-4.850) *** (-5.670) *** (-5.270) % VOL i *** (5.250) *** (5.500) *** (5.350) VCM i,post ** (-2.620) (-1.240) VPM i,post *** (-2.970) * (-1.650) --- VM i,post *** (-2.930) (-1.460) N R-square Panel B: PI i as the dependent variable Coef (t-stat) Coef (t-stat) Coef (t-stat) Coef (t-stat) Coef (t-stat) Coef (t-stat) Intercept (1.510) (0.16) (-0.070) (0.180) (0.300) (0.090) PI i,pre *** (-7.600) *** (-7.860) *** (-7.780) *** (-2.770) *** (-2.840) *** (-2.810) % VOL i * (1.920) * (-1.930) * (-1.940) VCM i,post (0.330) (1.240) VPM i,post ** (2.010) (1.340) --- VM i,post * (1.720) (1.170) N R-square Notes is the change in the speed of stock price adjustment to information after option listing; is the change in the pricing inefficiency after option listing, g i,pre and PI i,pre are the pre-listing speed of price adjustment to information and pricing inefficiency, respectively; % VOL i is the percentage change, between pre- and post-option listing, in average daily stock volume, VCM i,post, VPM i,post and VM i,post are the call option volume, put option volume and aggregate option volume, respectively; and N is the number options included in the regression. Note option volume measures are in logarithmic forms. ***, ** and * indicate significance at 1%, 5% and 10%, respectively. 17

18 References Amihud, Y., and Mendelson, H., (1987), Trading mechanisms and stock return: An empirical investigation, Journal of Finance 42, Atiase, R., 1985, Predisclosure information, market capitalization, and security price behaviour around earnings announcements, Journal of Accounting Research 23, Black, F., and Scholes, M., (1973), The pricing of options and corporate liabilities, Journal of Political Economy 81, Bollen, N., (1998), A note on the impact of options on stock return volatility, Journal of Banking and Finance 22, Cao, H., (1999), the effect of derivative assets on information acquisition and price behaviour in a dynamic rational expectation model, Review of Financial Studies 6, Cao, M., Wei, J., (2010), Option market liquidity: commonality and other characteristics, Journal of Financial Markets 13 (1), Cox, J., Rubinstein, M., (1985), Options markets, Englewood Cliffs: Prentice-Hall. Chelley-Steeley, P. L., (2008), Market quality changes in the London Stock Market, Journal of Banking and Finance 32, Chelley-Steeley, P. L., and Skvortsov, L., (2010), Efficiency and the trading system: The case of SETSmm, Journal of International Financial Markets Institutions and Money, doi: /j.intfin Chern, K., Tandon, K., Yu, S., and Webb, G., (2008), The information content of stock split announcements: Do options matter? Journal of Banking and Finance 32, DeTemple, J., and Jorion, P. (1990), Option listing and stock return: An empirical analysis, Journal of Finance 14, Easley, D., Hvidkjaer, S., O Hara, M., (2002), Is information-based risk a determinant of asset returns? Journal of Finance 52, Easley, D., O Hara, M., Srinivas, P., (1998), Option volume and stock prices: Evidence where informed traders trade, Journal of Finance 53, Faff, R., and Hiller, D., (2005), Complete markets, informed trading and equity option introductions, Journal of banking and Finance 29, Fedenia, M., and Grammatikos, T., (1992), Options trading and the bid-ask spread of the underlying stocks, Journal of Business 65, Grossman, S., (1988), An analysis of the implications for stock and futures price volatility of program trading and dynamic hedging strategies, Journal of Business 61, Hakansson, N., (1982), Changes in the financial market: Welfare and price effects and the basic theorems of value conservation, Journal of Finance 37, Harvey, A.C., (1992), Forecasting structural time series models and the Kalman filter, Cambridge University Press. Harris, L., (1989), S&P 500 Cash stock price volatilities, Journal of Finance 44, Jennings, R., and Starks, L., (1986), Earnings announcements, stock price adjustment, and the existence of option markets, Journal of Finance 41, John, K., Koticha, A., and Subrahmanyam, M., (1993), The microstructure of options markets: Informed trading, liquidity, volatility and efficiency, Working paper, New York University. Kumar, R., Sarin, A., and Shastri, K., (1998), The impact of options listing on the market quality of the underlying securities: An empirical analysis, Journal of Finance 53, Liu, S., (2009), The impacts of index options on the underlying stocks: The case of the S&P 100, The Quarterly Review of Economics and Finance 49,

19 Mazouz, K., (2004), The effect of CBOE option listing on the volatility of NYSE traded stocks: a time varying variance approach, Journal of Empirical Finance 11, Mayhew, S., and Mihov, V., (2004), How exchanges select stocks for option listing? Journal of Finance 50, Mendenhall, R., and Fehrs, D., (1999), Option listing and the stock-price response to earnings announcements, Journal of Accounting and Economics 27, Naik, V., and Lee, M., (1990), General equilibrium pricing of options on the market portfolio with discontinuous returns, Review of Financial Studies 3, Ni, S., Pan, J., and Pteshman, A., (2008), Volatility information trading in the option market, Journal of Finance 63, Pan, J., and Liu, J., (2003), Dynamic derivative strategies, Journal of Financial Economics 69, Roll, R., Schwartz, E., and Subrahmanyam, A., (2009), Options trading activity and firm valuation, Journal of Financial Economics 94, Roll, R., Schwartz, E., and Subrahmanyam, A., (2010), O/S: The relative trading activity in options and stock, Journal of Financial Economics 96, Ross, S., (1976), Options and efficiency, Quarterly Journal of Economics 90, Sahlstrom, P., (2001), The impact of stock option listings on return and risk characteristics in Finland, International Review of Financial Analysis 10, Sorescu, S., (2000), The effect of options on stock prices: 1973 to 1995, Journal of Finance 55, Stein, J., (1987), Information externalities and welfare reducing speculation, Journal of Political Economy 95,