Short Sales and Put Options: Where is the Bad News First Traded?

Transcription

1 Short Sales and Put Options: Where is the Bad News First Traded? Xiaoting Hao *, Natalia Piqueira ABSTRACT Although the literature provides strong evidence supporting the presence of informed trading in both the option and stock short market, it is not clear which market attracts more informed trading. Using a unique dataset that covers the intraday transaction data in the stock, option, and short market, we investigate informed trading in a cross-market environment, by explicitly studying the lead-lag relationship between the signed put volume and short sales of its underlying stock. Our high frequency analysis shows that in general short sales contain more information since it can predict subsequent stock and option returns and lead the net put volume. However, put options become more informative before the companies release their negative earnings announcements. Keywords: Put option, short sales, informed trading, earnings announcements This version: September 14, 2010 * C.T. Bauer College of Business, University of Houston, 334 Melcher Hall, Houston, Texas 77204, phone: , xhao2@mail.uh.edu. (Contact Author) C.T. Bauer College of Business, University of Houston, 334 Melcher Hall, Houston, Texas 77204, phone: , npiqueira@uh.edu. 1

2 Short Sales and Put Options: Where is the Bad News First Traded? This Version: September 14, Introduction Accumulating evidence in the literature supports the presence of informed trading in the option market. Black (1975) suggests that informed investors may choose to trade in option market since this market provides less transaction cost, capital requirements, trading restrictions, and higher leverage. A formal model developed by Easley, O Hara, and Srinivas (1998) suggests the existence of a pooling equilibrium in which informed investors choose to trade in both the stock and the option markets. Subsequent empirical research further documents support for the informed trading in option markets in the sense that option trading can predict subsequent returns for the underlying stock (Pan and Potesham (2006), Chakravarty, Gulen, and Mayhew (2004), Cao, Chen, and Griffin (2005), etc). At the same time, extensive recent research focuses on the role of short sellers in conveying valuable information about the stocks they short. Diamond and Verrechia (1987) suggest that short sellers should be informed since they cannot use the proceeds from short sales, and that market participants never short for liquidity reasons. Most of the empirical evidence suggests that short sellers are indeed informed traders and therefore, they have an important role in price discovery (e.g. Bohemer, Jones, and Zhang (2008), Diether, Lee, and Werner (2009), Bohemer and Wu (2009), among others). Although the literature provides evidence supporting the presence of informed trading in both the option and the stock short market, it is not clear which market attracts 2

3 more informed trading when negative information about the underlying stock is observed in the market. The purpose of this paper is to answer the above question by comparing the information content of short sales and put options trading. We study which market is more effective in the price discovery process when there is negative information about the underlying stocks. A comparison of the informational content between the short and put option market is important since it helps us to understand which market is more conductive to price discovery and information incorporation. The practical implications of this research are also relevant given the recent temporary changes in short selling regulation and the possibility of further changes in the near future. 1 If indeed short sellers play the most important role in contributing to price efficiency, what would be the impact for market efficiency of a more strict regulatory system? In particular, wouldn t informed short sellers attempt to trade on their information in different markets if short selling becomes too costly -- for example, the put option market? In this paper, we investigate the role of short sales and put option trading in revealing negative information about the stock, by explicitly studying the effects of short sales and put option trading imbalance on subsequent quote revisions and trading volumes in the short equity and put option markets. We use a unique dataset that covers intraday transaction data in the stock, option, and short equity market for a sample of NYSE stocks from January 2007 to June 2007 to estimate, using five minute intervals, a structural model based on Hasbrouck (1991). We extend his bivariate VAR model of 1 See, for example, Bohemer, Jones, and Zhang (2009) for an analysis of the effects of the September 2008 shorting ban for financial stocks. 3

4 stock market trades and quote revisions to also include the option market and the short equity market. By doing so, we can directly observe the price impact and the lead-lag relationships of put option trading and short sales, which allows us to compare the information contents between put option trading and short sales. Our high frequency analysis shows that during our sample periods short sales contain more information since they can predict subsequent stock and option returns and lead the net put volume, suggesting that more information can be learnt from the short equity market than from the put option market. We also perform subgroup analysis using different characteristics about the options and their underlying stocks in order to further compare the price discovery role between short sales and put option trading. The results suggest that, no matter which variable we use to group these stocks, short sales always contain more information since they have predictive power over subsequent put option returns and/or trading volume, providing strong support for the idea that informed investors will choose to trade in the short equity market first when they receive bad news about the stocks. The only different results are documented in our sub-period analysis, in which we study the trading in multiple markets within 7 days before a negative unexpected earnings announcement on the underlying stocks is released. With a pending event that will drive down the stock's fundamental value, more informed trading is observed in the put option market. This result suggests that the effect documented in Cao et al (2005), that call option plays a more important price discovery role before takeover announcement, can also be observed in the put option market. To our knowledge, our paper serves as the first attempt to use intraday data to investigate and compare informed trading in the put option and the short equity market in 4

5 a multi-market setting, contributing to our knowledge of how negative information is incorporated into the price system. Our results show that while the short equity market is important during normal periods in revealing negative information about the stocks, the option market may be more critical in the price discovery process before negative events occur, suggesting that both markets have their different roles in conveying information to the market. The remainder of this paper is organized as follows. Section 2 provides a brief review of related literature and our contribution. Section 3 describes the methodology. Section 4 describes the data and presents summary statistics. In Section 5 we discuss the main results. In Section 6 we present the results for different sub-samples and in section 7 we discuss our analysis before negative earnings announcements. Section 8 concludes. 2. Related Literature Our paper is related to two main fields of literature. First, we contribute to the literature investigating informed trading across different markets, in particular the studies comparing informed trading in the option and stock markets. Second, we contribute to the literature on informed short selling, in particular the studies linking return predictability and short selling activity. In this section, we briefly summarize the most relevant findings in each of these two fields and we present our contribution to the literature. Many empirical studies investigate the presence of informed trading in the options market and the role of option trading in revealing valuable information about the underlying stock. Black s (1975) argument that informed investors would prefer to trade options due to the higher leverage was tested by several subsequent studies. For example, 5

6 Easley, O Hara, and Srinivas (1998) find that informed trading also occurs in options markets. They propose and test a model that allows investors to choose whether they want to trade in stock or option market. They show that a pooling equilibrium i.e. an equilibrium in which informed traders trade on both the option and the stock markets -- will happen if the option market provides high leverage, or the liquidity in the stock market is low, or if there are more investors that are informed. Their empirical results show that signed option volume can predict future stock price movement, rejecting the separating equilibrium in which informed investors only trade in the stock market. Pan and Poteshman (2006) also provide evidence that option trading volume predict stock returns. More specifically, they find that stocks with the lowest put-call ratios (positive signal) outperform those with the highest put-call ratios (negative signal) in the near future (next day and next week). Chakravarty, Gulen, and Mayhew (2004) measure the price discovery in the stock and option markets by applying Hasbrouck s (1995) methodology to a sample of 60 firms, using intraday data. Their results provide evidence in favor of informed trading occurring in both stock and option markets since they find significant price discovery in the option market. Using intraday data, Cao, Chen, and Griffin (2005) study the information content of call options trading imbalance and underlying stock trading imbalance before takeovers. They find that during the preannouncement period, call options trading imbalance have higher predictability over the next-day stock returns (takeover premiums) than stock trading imbalance, providing further evidence on the presence of informed trading in the option market. The results, however are mixed when the focus is on which market (options or stocks) leads the other, i.e. where informed investors would trade first. Anthony (1988), 6

7 using total (not signed) daily volumes, studies the relationship between common stocks and their call option trading volume. His result shows that trading in call option can predict the trading in underlying stocks on the next day, i.e. options lead stocks. On the other hand, Chan, Chung, and Fong (2002) using a sample of 14 stocks, intraday data and a similar methodology used in our paper, find evidence that the stock market leads the option market, i.e. informed investors prefer to trade in the stock market (as in the separating equilibrium in Easley, O Hara, and Srinivas (1998)). Overall, the literature finds strong evidence supporting the presence of informed trading in the options markets but fails to form a consensus on which market (stocks or options) contains more information. Although most of the studies use signed volume in stock and option markets to compare their relative informativeness, none of them attempt to incorporate the trading volume from another market with strong informed investor presence the short equity market. Our paper aims to investigate and compare informed trading between the stock and option markets, by explicitly distinguishing the short sellers in the stock market. Our paper is also related to the literature on informed short selling, in particular the studies linking return predictability and short selling activity. Many empirical studies test the predictions of the theoretical model proposed by Diamond and Verrecchia (1987), in particular the implication that short sellers are informed. Most of the empirical evidence related to return predictability with monthly short interest data or intraday short trading (flow) data -- suggests that short sellers are informed traders and, by having 7

8 valuable bad news, are able to predict negative returns. 2 For example, Bohemer, Jones, and Zhang (2008) show that short sellers are well informed, by using proprietary daily short trading data from NYSE from 2000 to In particular they show that stocks that are heavily shorted underperform lightly shorted stocks by 1.16% (in risk-adjusted terms) in the subsequent 20 trading days. Diether, Lee, and Werner (2009) find that portfolios formed by buying lightly shorted stocks and selling heavily shorted stocks achieve positive abnormal returns, by using short sales intraday data in This is again, evidence that short sellers have valuable information (bad news) regarding the stock. Bohemer and Wu (2009) also use daily short trading data to show that short sellers increase the informational efficiency of stock prices, according to several efficiency measures. Using monthly short interest data, Asquith, Pathak, and Ritter (2005), show that stocks with high short interest underperform stocks with low short interest (only for equal-weighted portfolios). Desai et al. (2002) also show that abnormal negative returns are observed for stocks with higher monthly short interest. Although the empirical evidence summarized above suggests that short sellers are informed traders, these studies do not attempt to investigate short selling in a multimarket setting, in particular in an environment where informed traders might also choose to trade options. We contribute to this strand of literature by investigating the effects of short selling in a multi-market environment, aiming to shed light on how information flows across markets. Our unique dataset, which covers high frequency trading data in stock, option, and short markets, allows us to achieve this goal. To our knowledge, this is 2 In general, intraday shorting flow data is used to construct a daily series of short trading. 8

9 the first paper that uses intraday data to investigate the effects of both short selling and option trading on stock returns and subsequent trading volume. 3. Methodology and Empirical Predictions 3.1 Systems of Regression Equations for Multiple Markets The model we use to describe the dynamic relationship between trades and quote revisions in the stock, call, put and short markets is based on Hasbrouck (1991) and Chan et al. (2002). In Hasbrouck (1991), a bivariate VAR model of trades and quote revisions for the stock market is used to study the information content of stock trades, while in Chan et al. (2002), in order to compare the information roles of stock and option trades and quote revisions, this model is extended to include the option market. We further extend the structural model proposed by Chan et al. (2002) to also include trades in the short (stock) market. This is done by assuming that the information in short transactions cannot be fully conveyed from the imbalance of stock trading total volume. The basic bivariate VAR model for a single market is specified as follows: 3 r t = A(L)r t + B(0)z t + B(L)z t + ε 1,t (1) z t = C(L)r t + D(L)z t + ε 2,t (2) where r t is the quote revision at transaction time t, which is calculated as the change of bid-ask midpoint from these quotes following transaction t-1 to the quotes following transaction t, and z t is the total trading imbalance (positive if buy-initiated, and 3 See Hasbrouck (1991) for details on this specification (equations (2) and (4), pp ). 9

10 negative if sell-initiated) between transaction time t-1 and t. By assumption, the error terms in the two equations have zero means and are independent from each other. Chan et al. (2002) extend equations (1) and (2) to include trades and quotes in multiple (stock, call and put) markets. We follow the same line of reasoning to further include the trades in the short market into our structural model. Notice that the short market is part of the stock market, thus they share the same quote revision series and only trades in the short market are included in this model. More specifically, we define r t = r s,t r c,t r p,t and z t = z s,t z c,t z p,t z ss,t in equations (1) and (2), where r s,t, r c,t and r p,t represent quote revisions in the stock, call, and put market during time interval t, and z s,t, z c,t, z p,t and z ss,t represent net trade volume in the stock, call, put and short market during time interval t. A(L)(3 3), B(0) (3 4), B(L)(3 4), C(L)(4 3), and D(L)(4 4) (L=1, 2, 3) are coefficients to be estimated. Consistent with the methodology used in prior literature, lagged values of dependent variables on the right hand side are used to capture serial correlation effects, so that the disturbances can be assumed to be serially independent from each other. Compared to the system of two equations in Hasbrouck (1991) and the system of six equations in Chan et al. (2002), we have a system of seven equations in total. We believe that by explicitly distinguishing the short trading effects, we will be able to further investigate the role of short trading and (put) option trading in revealing information about the stock. 3.2 Empirical Predictions 10

11 Since the purpose of this paper is to compare the informational role of put option and short selling volume, we will focus on studying the effects of net put volume and short size on the subsequent return and trading volume in the put and short equity markets. First, with respect to the effects of net trade volume on subsequent quote revisions, we expect that net put trade volume (short size) should be more significant in predicting the subsequent stock and put market returns if there is more informed trading in the put (short) market. This methodology has been utilized in Chan et al. (2002), in which they compare the informational role of stock and option volume by studying their effects on subsequent returns in the market. In similar research, Cao et al. (2005) study the information content of call options by relating the option trading imbalance to the next day stock return. In the analysis of whether short sellers have valuable information about the stock, Diether at al. (2009) study the correlations between short sales and the stocks future returns. Second, we study the effects of net trade volume on subsequent net trade volume. We expect that if there is more informed trading in the short market (put option market), we should be able to observe an information flow from the short (put) market to the put option (short) market, thus short sales (net put option trade) should lead the subsequent put option trade (short sales). The lead-lag of trading volume in different markets has been studied in related research. For example, Anthony (1988) finds that trading in the call option market leads the trading of underlying shares with one day lag. He reasons that this represents an information flow from the option market to the stock market. It is also widely documented in the literature that option underwriters hedge their positions in 11

12 the stock market (Grundy et al. (2010), Battalio et al. (2010)). In this case, option trading is more informative and leads the stock trading. We also perform subsample analyses. Specifically, we divide the stocks according to their stock market and/or option market characteristics, and test whether these characteristics affect the informational role of short and put option trading. Since it is documented in the prior literature that call option trading tend to be more informed before corporate events (Cao at al. (2005)), we also conduct a sub-period analysis (7 days before the negative unexpected earnings is released) to see whether the same effect holds for put options. 4. Data and Summary Statistics The empirical analysis in this paper employs several different databases. The time-stamped option quotes and trades data are retrieved from the Bauer Research Dataset (BARDS) database. This dataset tracks and stores a continuous history of tickby-tick transaction data through Reuters for nearest-maturity options written on over 60 stocks that are listed on NYSE/AMEX and NASDAQ. 4 The stock market intra-day trading data come from the Trade and Quote (TAQ) database of NYSE, which provides a complete history of time-stamped quotes and trades data for all the underlying stocks whose options are tracked by BARDS. The intraday short sales data for these stocks also come from the TAQ database and are made available by NYSE as part of the requirements under Regulation SHO (January 2005). 5 For each 4 On each day, all the call and put options that have one month or less preceding their expiration are tracked. 5 For more information on Regulation SHO, check the website: 12

13 short-sale transaction, this dataset contains its transaction time, trade size, and an indicator that identifies short sales that are exempt from the Uptick Rule. 6 Our sample starts on January 22, 2007 and ends on June 8, Under Regulation SHO, approximately one third of the NYSE stocks (pilot stocks) are exempted from the Uptick Rule. Short sales of these pilot stocks cannot be observed by the market makers since they are masked by the NYSE display book software (Boehmer at al. (2008)). Only non-pilot stocks are kept in our analysis because according to Hasbrouck (1991), the VAR model adopted in this paper should be used when the trade innovations in different markets are observable so that they can have a price impact on related market. 7 Since the short-sale transaction dataset only coves stocks listed on NYSE, we are left with 27 non-pilot stocks whose options are also continuously tracked during our sample period in the BARDS dataset. We start our sample with all 27 stocks. Following Chan et al. (2002), each day the most actively traded put and call options are selected for each stock. When the most active option has five days or less to maturity, it is deleted from our sample since abnormal trading in the option market on the days near expiration are documented in prior literature. Thus each stock has at most 80 option days that can be used in this analysis. As for stock transaction data, only the trades and quotes that originate on the NYSE are included in our analysis since it is shown in Hasbrouck (1995) that the price discovery for NYSE stocks is more likely to take place on the NYSE rather than on other 6 A former rule established by the SEC that requires that every short sale transaction be entered at a price that is higher than the price of the previous trade. 7 For a complete list of pilot stock, see SEC' website: 13

14 exchanges. At the same time, since we need trading volume measurements over short (5 minutes) intervals, option days with thin trading are deleted from our sample. More specifically, following Chan et al. (2002), option trading days with fewer than 20 trades for the stock, the most active call, the most active put, or the short-sale are deleted. After that we are left with 729 option days in total. More than half of option days are deleted due to the thin trading problem in the option market. In order to classify the trading direction for each transaction on each stock and option in our sample, we use Lee and Ready (1991) algorithm. Specifically, if a trade is executed at a price above (below) the quote midpoint, it is classified as buy-initiated (sell-initiated). Trades at the quote midpoint are classified using the tick test, which determines the direction by comparing the trade price to the price of preceding trades. All the short-sales are classified as sell-initiated. Stock and option characteristics data are also collected for subgroup analysis. The Center for Research in Securities Prices (CRSP) data files are used to retrieved monthly and daily returns, market capitalization (defined as share price multiplied by the number of shares outstanding), share volume, and turnover (defined as share volume scaled by the number of shares outstanding). The number of analysts that make forecast for firms quarterly earnings is used to represent analysts coverage and retrieved from I/B/E/S. Monthly short interest data are directly purchased from authorized vendors of NYSE. Implied volatility and delta for the most active puts and calls are retrieved from Option Metrics. 14

15 Earning announcement data is collected from I/B/E/S for our analysis during preevent period. Unexpected earnings are defined as the difference between the actual earnings and the last estimated earnings for the same quarter. All the options days that are within 7 days before negative unexpected earnings announcement are chosen for our preevent studies. We present summary statistics in Table 1. The table reports the average daily trading volume for each stock in our sample (in number of shares traded), its most active call and put (in number of traded contracts, where each option contract is for one round lot of shares), and all of its calls and puts during the 729 sample option days. It also includes the average daily short volume of these stocks (in number of shares traded). Similar to previous literature, the average daily volume of the stock is larger than the average daily volume of its put and call options. Also, consistent with prior literature, the call option market is relatively more active than the put option market. On average the relative short volume accounts for 28% of the total stock trading volume. This is higher than the reported number (20%) in Boehmer and Wu (2009) possibly due to the sample difference, since we concentrate on a smaller group (27 stocks) of relatively larger firms. It should also be noticed that the trading volume of the most active call and put accounts for 45-98% of the trading volume of all calls and puts, which suggests that the most active call and put are good representatives for the option market of these stocks. With respect to signed trading volume, there is considerable variation in the percentage of buyinitiated and sell-initiated volume across stocks and options during our sample period. We continue our analysis of the sample properties in Table 2. This table provides a more detailed characterization of our sample, in particular with respect to trading 15

16 activity. First, we decompose total trading volume into average trade size and total number of trades. Consistent with prior literature, the daily number of trades in the stock market and the number of short sale transactions are much larger than the number of the trades in the corresponding option market. The quotation frequency, which represents the percentage of five-minute intervals having new quotes, is also much higher in the stock market, suggesting that the stock market is more liquid compared to the option market. This evidence indicates that illiquidity might be the major drawback that keeps an informed investor from trading in the option market, since she would be better off in the stock market if she wants to execute a large trade immediately. Finally, the option moneyness is also reported in Table 2 and, consistent with prior literature, the most active options appear to be at the money. 5. Main Results Five minutes intervals are used to estimate the structural model (1)-(2). Each option-day is partitioned into 78 successive five-minute intervals during the time period when both the option and the stock markets are open (from 9:30 A.M. to 4:00 P.M. EST). For each of these intervals, we generate five-minute quote revisions using the mid-point of the last bid and ask quotes for each of the stocks, the most active calls, and the most active puts. If no quote is available for an interval, there is no quote change for that interval, and we will keep the quote from last interval. The return of each interval is calculated as the log of the ratio of quote midpoints in successive intervals. We also calculate the net trade volume of the stocks and the most active calls and puts, and the 16

17 total volume of short-sales of these stocks for every five-minute interval. 8 Following Easley, O Hara, and Srinivas (1998) and Chan et al. (2002), we use the standardized return and net trade volume variables to control for the cross-section variations across different stocks and options. Each day we first calculate the mean and the standard deviation of the return and net trade volume for each variable. The variable is then standardized by subtracting the mean and dividing by the standard deviation. This allows us to pool the 729 option days for later analyses. Pool regression is used to estimate the structural model (1)-(2). Since we use standardized returns and net trade volume for each of variable in this model, we can assume that the error terms are homoscedastic. Furthermore, since we include lagged values of the dependent variables to control for serial correlation, we can also assume that the error terms are serially independent. The seven regression equations are estimated together using structural models, thus we control for the correlations between the equations. We include three lags for each explanatory variable, but for simplicity, we only report the first two lags in our results. The results of the main model are presented in Table 3 and we will discuss them in detail in the next two sub-sections. 5.1 Comparison with Chan et al. (2002) In this sub-section we compare our results in Table 3 -- relating options and the stock market -- with Chan et al. (2002). First, we confirm their result that the stock market net trading volume has predictive power over contemporaneous (coefficient is for calls and for puts) and future option returns (coefficient is for 8 Notice that all the short-sales are sell-oriented. 17

18 calls and for puts, for lag1), while the option net trading volume is only significant in explaining contemporaneous stock returns (0.025 and for call and put net trading volume, respectively). This is consistent with Chan et al. (2002) interpretation that stock market trading conveys more information than option market trading. Second, in terms of the relationship among returns we also find that, consistent with Chan et al. (2002), after controlling for net trading volume, quote returns in one market still have predictive power over subsequent quote returns in the other markets: for example, in the equation explaining stock returns, the coefficient for lag call returns is 0.013, and in the equation explaining call returns, the coefficient for lag stock returns is Within each individual market, a negative relationship between returns and their own lags are observed in both the stock and option markets, as in Chan et al. (2002). The main tests of our paper are (1) comparing the effect of short sales and net put trading on the subsequent quote revisions in the two markets, and (2) studying the leadlag relationship between short sales and net put trading. The purpose of these analyses is to compare the relative price discovery role of the short and put markets. We will discuss them in detail in the next sub-section. 5.2 The informational role of short sales and put option trading First, with respect to the effects of short sales and net put trade volume on the subsequent quote revisions in these two markets, Table 3 suggests that only short sales have predictive power over both future stock and option returns. In particular, an increase in short sales is positively (negatively) related with subsequent put (call) returns. We also show in Table 3 that short sales have predictive power over stock market returns in 18

19 subsequent 5-minute intervals (e.g. coefficient is -0.03, with a t-stat of for lagged short size, lag 1), indicating again that short selling possibly conveys important negative information regarding the stock, which is not necessarily revealed by looking only at the stock net trading volume. Notice that it has been well documented in the literature that short sellers are able to spot overvalued stocks and their trading predict future stock returns (e.g. Diether et al. (2009), Asquith, Pathak, and Ritter (2005), and Bohemer and Wu (2009)). Our finding, however, is different from previous literature since it is the first study to use intraday data (with 5-minute intervals) to study the information content of short sales in a multi-market setting. The same results cannot be found for the net put trade volume. As discussed earlier, net put option trading is only significant in explaining its own subsequent quote revisions. This result suggests that there is more information contained in short sales in the equity market, compared to put option trading. Second, about the relationship between the net trading volume of short and put markets, we predict that if the put market (short market) has leading information, put net trading volume (short sales) should have predictive power over subsequent short sales (put net trading volume). If the trades in both markets contain different information and are unrelated with each other, we should not observe the lead-lag relationship between trading in the two markets. Table 3 provides the main results regarding the lead-lag relationship between trading in different markets. First we notice that, consistent with Chan et al. (2002), the three net trading volume series (stock, put and call) are not significantly cross-correlated. However, once we include short sales in the regression, we observe significant explanatory power of short sales on subsequent put net trading volume (coefficient is with a t-stat=3.32), while put net trading volume is not 19

20 significant in explaining future short sales. This indicates that there is more informed trading in the short market, and a possible information flow from the short to the put option market. Both the analyses of the price impacts of put and short trading and the lead-lag relationships between the two market indicate that compared to put option, short sale volume may be more informative. Notice that prior literature suggests that informed investors will prefer to trade options because the higher leverage offered by this instrument (Black (1975)). However, our results suggest that during normal market time, short sale trading volume is a better predictor for subsequent stock and put option returns. 6. Subsample Analysis To further compare the informational role of put option trading and short sales, we divide the 27 stocks into two groups according to each stock and/or option characteristic that might affect informed investors decision as which market to trade, and then apply the structural model to each group to compare their estimation results. 6.1 Analysis based on stock characteristics First, stock market-related characteristics are used in order to perform subgroup analysis. These characteristics include the exempt conditions of short sales, firm size, share turnover, past returns, price, short interest ratio, and number of analysts following these stocks. Table 4 presents the estimation results. As discussed in Boehmer and Wu (2009), short sales that are exempt from the Uptick Rule are less likely to be information motivated, because they mainly result from 20

21 market making activities. Following this logic, we first separate all the short sales in our sample into exempt and nonexempt groups, and then apply each of them to the structural model in our main analysis (Table 4, Panel A). It is expected that the leading role of short market mainly comes from all the non-exempt short sales. Consistent with our expectation, first we find that for the short sales exempt from Uptick rule, even though they can still predict subsequent stock return, they are no longer significant in predicting subsequent put returns. But the non-exempt short sales have predictive power for the future returns in both stock and put market. Second, the short market is leading only for the non-exempt short sales group in our regression, while put option trading is marginally leading if we use the exempt short sales in our regression. In all, our results suggest that non-exempt short sales contain more information and explain the leading position of short sales documented in the last section. Firm size is then used to group stocks. Since larger stocks are normally held by institutions, it is easier for informed investors to generate short positions for these stocks, thus more information can be traded through short sales. 9 On the other hand, it is reasonable to expect that larger firms would have a more liquid option market, thus allowing more informed trading through put options. Thus, larger firms motivate informed trading in both short and put option market, and it is interesting to compare which market has leading information for these firms. Table 4 (Panel B) shows that for larger firms, short sales can still predict subsequent stock and put returns, and lead the put option trading, suggesting that even when both option and short market are relatively active, informed investors will be more likely to trade in the short market first. For 9 For example, D Avolio (2004) notes that institutional investors holding long positions provide the majority of shares for short sales. 21

22 smaller firms we cannot make conclusions as which market s informed trading is more intense, since in this case short sales can only predict future stock returns while put option only predict subsequent put returns. Meanwhile, trading in the two markets is also uncorrelated with each other. This might result from the fact that both the option and the short markets are not very active for these firms due to liquidity reasons, or to the fact that informed trading occurs in both markets. Our analysis does not allow us to differentiate between these two hypotheses. Stock market turnover (Table 4, Panel C) is also used in the subgroup analysis. We find that short sales predict subsequent stock and put option returns, and lead put option trade for both high and low turnover stocks, indicating that our sample turnover is possibly unrelated with the relative informed trading in the two markets. Average six-month stock returns (Table 4, panel D) during our sample period are then used to group the stocks. Since stocks with lower returns are more likely to be overvalued during our sample period, we expect to find a stronger leading position of short sales among this group of stocks. Consistent with our expectation, first we find that short sales are more economically and statistically significant in predicting future put returns for stocks with lower returns. Second, short sales only lead net put option trade for stocks with low returns during our sample period, while trades in the short and put option market are unrelated with each other for stocks with higher returns. This further supports our finding that when a stock is overvalued, insiders are more likely to first trade in the short market. 22

23 The argument for stock price (Table 4, Panel E) is similar to those for firm size. It is reasonable to believe that stocks with higher price will have a more liquid option market and are more likely to be held by institutions -- thus it is also easier for informed investors to sell short these stocks (D Avolio (2004)) Similar results to the firm size subgroup analysis are found here. Even though short sales can predict future returns in stock and option market for both high and low price stocks, they only lead net put option trade for stocks with higher prices. No lead-lag relationship is found for stocks with low prices. Again, this result confirms our finding that, in general, informed investors choose to trade first in the short market. The average short interest ratio (SIR) for each stock during our sample period is also used to perform subgroup analysis. It has been documented in the literature (Ofek et al. (2004), Reed (2007)) that low short interest ratio indicates higher short sale constraints. Thus, we expect to find that at least part of the informed investors who face short sale constraints will turn to the put option market if they think the stock is overvalued. Our results (Table 4, Panel F) provide support for this argument. Even though short sales are able to predict future stock and put option returns for both low and high SIR stocks, for stocks with lower SIR (higher short sale constraint), both short sales and put option trading can predict each other s subsequent trading volume. This result suggests that for stocks with higher short sale constraint, some informed trading will occur in the put option market. The last stock-related characteristic is the number of analysts following each quarter s stock earnings announcement. Since less private information can be uncovered for stocks with high analyst coverage (Easley, O Hara, and Paperman (1998)), it is 23

24 expected that there will be less private information-induced trading for these stocks. However, as pointed out by Roll et al. (2009), if institutional clients have access to relevant private information through a larger number of analysts (Green (2006)), these clients are more likely to exploit this (negative) information in the short or put option markets. Thus the impact of analysts coverage on the relative informed trading in the two markets is ambiguous. Table 4 (Panel G) shows that short sales predict subsequent stock and put option returns and lead net put option trade for stocks with both high and low analyst coverage, suggesting that the number of analysts is unlikely to be related with the relative informed trading in the two markets. 6.2 Analysis based on option characteristics In this sub-section, option-related characteristics are used to separate the sample stocks. First, put options with higher delta (less negative) are more likely to be traded for hedging needs, as explained in Roll et al. (2009). This is because a more negative put option delta indicates a higher sensitivity to the price change of underlying stocks, and options with lower deltas (larger absolute magnitude) require fewer option contracts per underlying stock share to achieve the same share-equivalent position. Thus, it is expected that short sales will still be more informed than options with higher delta, since the option trading for these options is more likely driven by hedging purposes, not information. Results in Table 5 support this expectation. For options with lower deltas, both the short sale and put option have predictive power for each other s subsequent trading, indicating that in this case, some information is first traded in short market, and some in put option market. 24

25 Implied volatility is also expected to be related with option trading because higher implied volatility indicates higher value for option trading. However, the relationship between informed trading and implied volatility is not clear because it can attract both informed and uninformed investors. Results in Table 5 shows that short sales contain more information than put option trading for options with both high and low implied volatility; while this effect is more significant for options with high implied volatility, suggesting that informed investors will choose to trade in short market first even if the underlying stocks have higher implied volatility. 6.3 Analysis based on stock and option characteristics In this section, we combine the characteristics of the options and underlying stocks. As discussed above, one of the drawbacks of choosing to trade in the option market is the option market s relative illiquidity. In order to address this issue, we group the options and underlying stocks in our sample according to the relative liquidity in the put option and (short) stock market. First we calculate the turnover put-short ratio for each stock by dividing put option turnover by total short sales volume during our sample period (as in Roll and Subrahmanyam (2009)). Higher put-short ratio indicates higher liquidity in the put option market relative to the stock market. It is not clear now the relative liquidity in the short and put option markets is going to affect the informed trading in these two markets, since on one hand, in a more liquid option market, investors face less transaction costs related with illiquidity; on the other hand, liquid option market may also indicate high hedging or uninformed trading. Table 6 (Panel A) suggests that for stocks with both high and low 25

26 put-short turnover ratio, put option trading is not significant at predicting future stock market returns or subsequent short sales, while short sales are always significant in predicting future returns and trading volumes for both high and low turnover put-short ratio stocks. Thus, we conclude that higher turnover in the put option market cannot indicate more informed trading in this market. The second liquidity measure we use is bid-ask spread. We construct put-short spread ratio by dividing the closing bid-ask spread in the option market by the closing bid-ask spread in the stock market, thus higher put-short spread ratio represents higher liquidity in the put option market compared to the stock market. According to Table 6 (Panel B), results are qualitatively similar when the second liquidity measure is employed except that put option trading is significant at predicting subsequent stock returns at p<0.10. For stocks with both high and low put-short spread ratio, put option trading is not significant at predicting subsequent short sale volume. However, short sales lead the option trading even after we control for the option and stock market liquidity, suggesting that short sales are more informative compared to put option trading regardless of the relative liquidity of short equity and put option market. In all, to further understand the different informational roles of short sales and put option trading, we perform the subgroup analysis using different characteristics about the options and their underlying stocks. Our results suggest that short sales play a more important price discovery role than put option trading when there is negative information about the underlying stock. 7. Sub-period Analysis 26

27 Prior literature (Cao et al. (2005)) finds that call option trading contains more information before company takeovers in the sense that during the preannouncement period, call option trading imbalance has great predictability over the next-day stock returns (takeover premiums) than stock trading imbalance; while the same effect cannot be found during non-takeover period. In particular, they find that prior to announcements, buying activity is highest in the short-term out-of-the-money call options (with the highest leverage). This suggests that the leverage advantage in option markets is more significant and can attract informed traders when there is a pending event that will change the company s fundamental value. Thus, in this section, we also conduct pre-event analysis to test whether the same effect can be found between put option trading imbalance and short sales before a company s negative earnings announcement is released. We check all the quarterly earnings announcements of our 27 stocks during the sample period of time and calculate their unexpected earnings (the difference between actual and predicted earnings). There are 11 negative earnings announcements in total. For each of them, we keep 7 days before the announcement day and apply our structural model to the 77 option days. We expect to find that during this pre-announcement period, put option trading should contain more information in the sense that it should be more significant in predicting future stock market returns and short sales volume. The results in Table 7 are consistent with our expectation. During the 7 days before negative earnings announcement is released, put option trading imbalance is significant in predicting subsequent stock market returns and close to significant in predicting future short sales at p<0.10. Moreover, we find that short sales are no longer 27

28 significant in predicting subsequent returns and put option trading volume, suggesting that in a pre-event period, put option trading contains more information in the sense that it is a better predictor for subsequent stock market returns. Our results indicate that the effect documented in Cao et al (2005) can also be found in the put option market. Cao et al. (2005) show that with a pending extreme informational event (takeover), the option market plays an important role in the price discovery process while normally stock market trading imbalances are predictors of next day stock returns and option volume is uninformative. Our results suggest that while during normal market time there is more informed trading in the short market, during the pre-event period, informed investors will turn to the put option market and the put option trading is more informative. However, we also acknowledge that since our sample is relatively small (11 announcements in total), our results should be interpreted with caution. Our main results still suggest that short sales play a more important price discovery role than put option trading when there is negative information about the underlying stock. 8. Conclusion This study provides an empirical comparison of the price discovery role between net put trading volume (buyer-initiated trading volume minus seller-initiated trading volume) and short sales of the underlying stocks for a group of actively traded NYSE stocks and their corresponding put options. It has been documented in the literature that both put option trading and short sales convey valuable negative information about the underlying stocks. However, it is not clear which market attracts more informed investors. 28