Joint Effects of Financial Transparency and Media Coverage in. Contrarian Stock Market

Transcription

1 Joint Effects of Financial Transparency and Media Coverage in Contrarian Stock Market Hung-Wen Lin 1, Jing-Bo Huang 2, Kun-Ben Lin 3, Shu-Heng Chen 4 * Abstract: We study financial transparency and media coverage in the Chinese stock markets. A symmetric solution is found in this paper. Under the high transparency and high media coverage portfolios, the market has a positive profit from conditional momentum (See Avramov et al., 2007; Menkhoff et al., 2012). It has a negative profit from break-down momentum using low transparency and low media coverage portfolios (See Grinblatta and Han, 2005; Asem, 2009). There is a very high percentage of retail investors in the Chinese stock market. Above outcomes show that the spread speed of news and amount of information are important in this market. Keywords: Transparency, Media Coverage, Momentum, Contrarian 1. Introduction The Chinese stock markets have developed a lot in the past twenty years. The development has made the listed firms pay more attention to their financial report quality. The regulatory authority also supervises the financial operation of firms a lot. As the Chinese society develops, media gradually plays a more and more important role in information transmission. In China, newspaper, television, internet and instant communication tools strongly spring up. It is obvious that financial transparency and media coverage are important in China. Hence, it is necessary to study the effect of financial transparency and media coverage in this emerging market. Our study also derives from the effect of transparency and media coverage on 1 Nanfang College of Sun Yat-sen University 2 Sun Yat-sen University 3 Nanfang College of Sun Yat-sen University 4 National Chengchi University * Corresponding Author chen.shuheng@gmail.com 1

2 financial markets. For instance, transparency will affect stock prices and stock returns. Bleck and Liu (2007) have found opacity strengthens the asset price crashes both in magnitude and frequency in different markets. Dasgupta et al. (2010) found that stock return synchronicity and transparency are positively correlated in America. In addition, Liao et al. (2011) found that the abnormal return and transparency are positively correlated. The investors are more likely to sell the less transparent stocks. Media coverage has effects on stock prices. Aman (2013) also found that price crashes frequency increases with media coverage. Kim et al. (2016) pointed out that media coverage and stock price synchronicity are negatively correlated in China. The content of media report will affect stock price. Birz (2017) found that the stock price and media coverage are tightly correlated. The tones in the news and stock price are positively correlated. Some literatures study momentum or contrarian from the perspective of information, such as Chan et al. (1996), Johnson (2002), Kang et al. (2002), Da et al. (2014), Jiang et al. (2005) and Andrei and Cujean (2017). Transparency and media coverage also have effect on information. According to Bushman et al. (2004), transparency is the availability of firm specific information. Based on this definition, we think that transparency implies the firm specific information amount. Media is the essential tool for information transmission in modern society, so it indicates information spread. According to Jegadeesh and Titman (1993), we find significant contrarian in Chinese stock markets. This phenomenon is in line with other studies of Chinese stock markets, such as Wang (2004), Wang and Chin (2004), Cheema and Nartea (2014) and Cheema and Nartea (2017), etc. However, it is inconsistent with the studies of other developed stock markets, such as Rouwenhorst (1998), Hong et al. (2000), Lesmond et al. (2002), Griffin et al. (2003), Korajczyk and Sadka (2004), Cooper et al. (2004) and Hillert et al. (2014), etc. We want to study the effect of transparency and media coverage in this contrarian stock market. We construct some two-way sorted portfolios to study the individual effect of transparency and media coverage in this market. We sort the stocks by stock return and transparency (media coverage) with different sorting orders in the two-way 2

3 sorted portfolios. Some three-way sorted portfolios are constructed to study the joint effects of transparency and media coverage. We sort the stocks by transparency, media coverage and stock return with different sorting orders in the three-way sorted portfolios. Finally, we study whether the joint effects of transparency and media coverage are robust or not. We extend the formation period and holding period of our three-way sorted portfolios. The remainders of this paper are as follows. Section 2 introduces our models and data. Section 3 shows the results of our empirical analysis. Section 4 contains our robustness checks. The conclusions of this paper are in Section Models and Data This section includes the computations of transparency, media coverage, and momentum and data description. Transparency measure contains earnings aggressiveness, earnings smoothing and loss avoidance. We do not show the details of each model due to the length of this article. 2.1 Integrated Transparency and Media Coverage Measure Following Bhattacharya et al. (2003), we compute earning aggressiveness, earning smoothing and loss avoidance of each stock. Decile rankings are applied to these three variables and the average decile of each stock is the integrated transparency. We use the modified Jones model from Dechow et al. (1995) to measure earning aggressiveness. Our measure for earnings smoothing follows the methodology of Mclnnis (2010). Following the methodology of Ball and Shivakumar (2005), we use the timeliness of loss recognition to be the proxy of loss avoidance. Following Hillert et al. (2014), we use the residuals of the regression model as media coverage. Additionally, we make some adjustments regarding the Chinese market. 2.2 Momentum Calculations According to Jegadeesh and Titman (1993), our momentum calculations are as follows. We use J to represent the past J-season formation period and K is the future K- season holding period. In each period, we use the formation stock returns to select stocks. The stocks with highest formation returns are the winner portfolio, while the 3

4 stocks with lowest formation returns are the loser portfolio. In holding period, we compute the return difference between winner portfolio and loser portfolio (winner minus loser). The total average of return difference is the momentum profit. 2.3 Data Description We collect the data of stock price and financial transparency from the China Stock Markets and Accounting Research (CSMAR) database. The data of media (articles of stocks) is from the China Infobank database. The data is from 2005 to 2016, as CSI300 index started from It contains both A-share stocks and B-share stocks in the Chinese stock markets, including Shenzhen Stock Exchange and Shanghai Stock Exchange. We delete the stocks without complete data. Following Jegadeesh and Titman (1993), we sort all the stocks in the whole market by stock return to select winner stocks and loser stocks. The formation period and holding period are two seasons. The total mean value of momentum profit is significantly negative (t-stat is ), implying there is contrarian in China. 3. Empirical Analysis Except stock return, we still have transparency and media coverage. Under the definition of Jegadeesh and Titman (1993), the contrarian of Chinese stock market has been documented in many studies, such as Wang (2004), Wu (2011) and Lin et al. (2018), etc. Thus, we would like to study the effect of transparency and media coverage in this market by sorting stocks. Based on the above discussions, we construct several two-way sorted portfolios to dissect the individual effect of transparency and media coverage. We also construct several three-way sorted portfolios to detect the joint effects of transparency and media coverage. 3.1 Two-way Sorted Momentum Portfolios Bushman et al. (2004) defines transparency as the availability of the firm specific information. Transparency indicates the information amount of the listed firms. What 4

5 is the individual effect of transparency in the contrarian market? We construct several two-way sorted momentum portfolios to study it. As the information transmission tool in modern society, media implies information spread. Fang and Peress (2009) found that the stocks with less media coverage or without media coverage have higher return than those with more media coverage in America. Their findings suggest that media coverage has an important effect on stock return. What is the individual effect of media coverage in the contrarian market? Some two-way sorted portfolios are constructed to study it. Table II Two-way Sorted Momentum Portfolios This table shows the outcomes of two-way sorted momentum portfolios. We sort the stocks by transparency into three groups including high transparency, middle transparency and low transparency (HT, MT, LT). The sorts of media coverage are also three groups including high media coverage, middle media coverage and low media coverage (HC, MC, LC). Transparency and media coverage are called as additional sorting factor (asf). We sort the stocks into five groups by stock return to select winner stocks and loser stocks with ascending order. The formation period and holding period are two seasons. Interacted momentum portfolios sort all the stocks in the whole market by stock return and additional sorting factors independently. The intersections are what we need. Two interacted momentum portfolios are created. The first strategy is created by transparency and stock return and it is noted as Inter (R, T). The second strategy is created by media coverage and stock return and it is noted as Inter (R, C). Break-down momentum portfolios first sort all the stocks in the whole market by stock return to construct winner portfolio and loser portfolio. In winner portfolio and loser portfolio, the stocks are sorted by transparency or media coverage. We have two break-down momentum portfolios. The strategy created by stock return and transparency is noted as Seq (R, T). The strategy created by stock return and media coverage is noted as Seq (R, C). Conditional momentum portfolios first sort all the stocks in the whole market by transparency or media coverage into several groups. In each group, we use stock return to sort the stocks and select the winner stocks and loser stocks. We finally have two conditional momentum portfolios. Transparency and stock return make Seq (T, R) strategy. Media coverage and stock return make Seq (C, R) strategy. *, **, *** represent significance at 10%, 5% and 1% respectively. Panel A: Inter (R, T) Momentum Panel B: Inter (R, C) Momentum MP t-stat MP t-stat HT *** HC MT MC LT * LC * Panel C: Seq (R, T) Momentum Panel D: Seq (R, C) Momentum MP t-stat MP t-stat 5

6 HT *** HC MT MC *** LT ** LC Panel E: Seq (T, R) Momentum Panel F: Seq (C, R) Momentum MP t-stat MP t-stat HT HC MT MC LT LC Inter (R, T) strategy makes negative momentum profits significantly with high transparency and low transparency (t-stats are and , respectively). Seq (R, T) strategy also makes negative momentum profits significantly with high transparency and low transparency (t-stats are and , respectively). These phenomena suggest that the transparency does not have its individual effect in the market. Winner stocks and loser stocks will not maintain their past price trend by effect of information amount. With low media coverage, Inter (R, C) strategy has a negative momentum profit significantly (t-stat is -1.68). In addition, Seq (R, C) strategy also has a negative momentum profit significantly (t-stat is ) with middle media coverage. These findings are similar with transparency. Media coverage does not have its individual effect in the market. Whether the information spread is wide or not, winner stocks and loser stocks cannot maintain their past price trend. Hillert et al. (2014) think that the more coverage significantly presents stronger momentum. The results also reveal that the conclusion of Hillert et al. (2014) is not suitable for Chinese market. More coverage in media may not make momentum in China. 3.2 Three-way Sorted Momentum Portfolios For the sake of studying the joint effects of transparency and media coverage in the contrarian market, we make several three-way sorted momentum portfolios. The portfolios are produced from different sorts with diverse orders. Table III Three-way Sorted Momentum Portfolios 6

7 This table shows the outcomes of three-way sorted momentum portfolios. We sort the stocks by transparency into three groups including high transparency, middle transparency and low transparency (HT, MT, LT). The sorts of media coverage are also three groups including high media coverage, middle media coverage and low media coverage (HC, MC, LC). Transparency and media coverage are called as additional sorting factor (asf). We sort the stocks into five groups by stock return to select winner stocks and loser stocks with ascending order. The formation period and holding period are two seasons. Interacted momentum portfolios sort all the stocks in the whole market by stock return and additional sorting factors independently. The intersections are what we need. One interacted momentum strategy is created. The strategy created by stock return, transparency and media coverage is noted as Inter (R, C, T). Break-down momentum portfolios first sort all the stocks in the whole market by stock return to construct winner portfolio and loser portfolio. In winner portfolio and loser portfolio, the stocks are sorted by transparency and media coverage with specific orders. We have two break-down momentum portfolios. The strategy created by stock return, transparency and media coverage is noted as Seq (R, T, C). The strategy created by stock return, media coverage and transparency is noted as Seq (R, C, T). Conditional momentum portfolios first sort all the stocks in the whole market by transparency and media coverage with specific orders into several groups. In each group, we use stock return to sort the stocks and select the winner stocks and loser stocks. We finally have two conditional momentum portfolios. Transparency, media coverage and stock return make Seq (T, C, R) strategy. Media coverage, transparency and stock return make Seq (C, T, R) strategy. *, **, *** represent significance at 10%, 5% and 1% respectively. Panel A: Inter (R, C, T) Momentum Panel B: Seq (R, T, C) Momentum MP t-stat MP t-stat HC & HT HT & HC HC & MT HT & MC HC & LT HT & LC MC & HT MT & HC ** MC & MT * MT & MC * MC & LT MT & LC LC & HT LT & HC LC & MT LT & MC LC & LT ** LT & LC ** Panel C: Seq (R, C, T) Momentum Panel D: Seq (T, C, R) Momentum MP t-stat MP t-stat HC & HT * HT & HC * HC & MT HT & MC HC & LT HT & LC MC & HT MT & HC MC & MT MT & MC MC & LT MT & LC LC & HT LT & HC LC & MT LT & MC

8 LC & LT * LT & LC Panel E: Seq (C, T, R) Momentum MP t-stat HC & HT * HC & MT HC & LT MC & HT MC & MT MC & LT LC & HT LC & MT LC & LT * Inter (R, C, T) strategy, Seq (R, T, C) strategy and Seq (R, C, T) strategy make negative momentum profits significantly (t-stats are , and respectively) with low transparency and low media coverage. It means that these portfolios exhibit contrarian by low transparency and low media coverage. However, Seq (T, C, R) strategy and Seq (C, T, R) strategy have positive momentum profits significantly (t-stats are and respectively) under high transparency and high media coverage. High transparency and high media coverage enable these portfolios to exhibit momentum. High transparency means high information availability. This circumstance enables the investors to acquire much information of the firms. The investors are able to assess the operation status of the firms. And high media coverage also lets information spread widely. Savor (2012) found that the zero-investment portfolios of buying winner and selling loser will significantly make positive profits in the information-based circumstance, implying that information portfolio exhibits momentum. In addition, Bushman et al. (2004) found that transparency and media coverage are positively correlated. Solomon et al. (2014) found that media coverage will urge the investors to pay more attention to the past stock return. As a result of high transparency and high media coverage, the stocks will be extremely covered in the media and with much information. The decisions of investors are possibly consistent with actual situation. 8

9 The consensus of investors tend to be strengthened and maintained by high transparency and high media coverage. The investors buy winner stocks and sell loser stocks, which makes the price of winner stocks continue to increase and the price of loser stocks continue to decrease in the future. Thus, momentum emerges. Low transparency makes low information availability. This circumstance lets the investors acquire little information. The investors cannot assess the operation status of firms. Low media coverage does not let the information spread widely. Savor (2012) also found that selling winner stocks and buying loser stocks significantly make positive profits in the circumstance of no information, implying that no-information portfolio has contrarian. Chan (2003) found that investors will overreact to the information and the price changes will face price reversal in the circumstance without news headlines. The impact of low transparency and low media coverage will easily lead the decisions of investors to be contradictory to the actual situation. When the actual situation is out of the expectations of investors, the investors will overreact to the information and change their decisions. The price changes stemming from the trading activities will meet price reversal. Therefore, contrarian arises. 4. Robustness Checks In order to confirm whether the findings in section 3 are robust or not, we also extend the formation and holding period of Seq (R, C, T) strategy, Seq (R, T, C) strategy, Seq (T, C, R) strategy and Seq (C, T, R) strategy. 4.1 Extensive Formation and Holding Period We extend the formation period and holding period to confirm whether the joint effects of transparency and media coverage are robust or not. The formation period and holding period are extended from two seasons to twelve seasons High Transparency and High Media Coverage We study the joint effect of high transparency and high media coverage in the extensive horizons. The formation period and holding period are from two seasons to 9

10 twelve seasons. The portfolios are Seq (R, T, C) strategy, Seq (R, C, T) strategy, Seq (T, C, R) strategy and Seq (C, T, R) strategy. With high transparency and high media coverage, the most of extensive momentum profits are negative. Even if there are some positive momentum profits, the most of them are insignificant or close to 0. For instance, with high transparency and high media coverage, Seq (R, T, C) strategy insignificantly has a positive momentum profit of when the formation is two seasons and the holding is three seasons (t-stat is 0.243). Moreover, under high transparency and high media coverage, Seq (C, T, R) strategy insignificantly has a positive momentum profit of with four-season formation and three-season holding (t-stat is 0.798). The joint effect of high transparency and high media coverage are easily changed. In our original outcomes, we document that high transparency and high media coverage exhibit momentum with horizons of two-season formation and two-season holding. It is evident that high transparency and high media coverage exhibit contrarian when we extend the formation and holding. The results suggest the information may be fake and has continuation in China. Kang et al. (2002) found that Chinese stock market is lack of reliable information. Fan et al. (2013) similarly found that the managers always mask the true economic performance of firms in China. However, the investors fail to identify the truth of information and regard the fake information as true by mistake in this circumstance. Yang and Luo (2014) also found that investors are unable to distinguish the true information between false information in China. The past price trend of winners and losers are maintained. Hence, momentum emerges temporarily during two seasons. As the time goes by, the true information is revealed and the price is corrected, which causes price reversal. Finally, this situation leads to contrarian Low Transparency and Low Media Coverage 10

11 What is the joint effect of low transparency and low media coverage in the extensive horizons? We extend the formation period and holding period from two seasons to twelve seasons. The portfolios are Seq (R, T, C) strategy, Seq (R, C, T) strategy, Seq (T, C, R) strategy and Seq (C, T, R) strategy. The joint effect of low transparency and low media coverage makes the most of extensive momentum profits are also negative. Some positive momentum profits exist, but a lot of them are insignificant or close to 0. For instance, with low transparency and low media coverage, Seq (R, C, T) strategy does not have positive momentum profits regardless of the length of formation and holding. Additionally, under low transparency and low media coverage, Seq (T, C, R) strategy insignificantly has a positive momentum profit of with six-season formation and two-season holding (t-stat is 0.112). Based on the above outcomes, we can see that the joint effect of low transparency and low media coverage is robust in this contrarian market. It will not be changed in any horizons of formation and holding. 5. Conclusions From the perspective of information, we find that transparency and media coverage have their important effects in Chinese stock market. They help us to understand information amount, information spread, the quality of corporate information and investor behaviors in this market. The transparency or media coverage does not have individual effect in the market. We document this result in two-way sorted momentum portfolios. Thus, we study the joint effects of transparency and media coverage in the contrarian market. High transparency and high media coverage make momentum, whereas low transparency and low media coverage still make contrarian. They also imply that the information amount and information spread are essential in the contrarian market. These outcomes are basically consistent with Chan (2003), Savor (2012) and Solomon et al. (2014). High transparency and high media coverage enable the investors to assess the 11

12 operation status of firms. The information spreads widely. The decisions of the investors in this circumstance are easily consistent with actual situation. The investors buy winners and sell losers, which lets the price of winner stocks continues to increase and the price of loser stocks continues to decrease. Hence, momentum emerges. However, this phenomenon turns to contrarian when we extend formation and holding. We think the information is fake but it is regarded as true by mistake and has continuation. This idea is consistent with Kang et al. (2002), Fan et al. (2013) and Yang and Luo (2014). With high transparency and high media coverage, the fake information is strengthened and make investors have the same opinions. Therefore, this circumstance maintains the past price trend and momentum temporarily emerges. As time goes by, the true information is gradually revealed. It makes the price be corrected and price reversal. Therefore, contrarian arises. With low transparency and low media coverage, the investors have difficulties in assessing the operation status of firms and the information just spreads in a small scale. The decisions of the investors in this circumstance are possibly contradictory to the actual situation. The investors will overreact to the information and change their decisions, so the past price trend of winners and losers cannot continue. Finally, contrarian arises. No matter how we extend formation and holding, this phenomenon stays. Compared to other developed markets, the Chinese stock markets are special. The Chinese markets have significant contrarian profits, suggesting that price reversal always happens and the markets have extreme risks. We think momentum and risk studies deserve a lot of discussions. This is left for future research. 12

13 Reference Aman, 2013, An analysis of the impact of media coverage on stock price crashes and jumps: Evidence from Japan, Pacific-Basin Finance Journal, Anctil, Dickhaut, Kanodia and Shapiro, 2003, Information Transparency and Coordination Failure: Theory and Experiment, Journal of Accounting Research 42, Andrei and Cujean, 2017, Information percolation, momentum and reversal, Journal of Financial Economics, Accepted Manuscript. Asem E., 2009, Dividends and price momentum, Journal of Banking & Finance 33, Avramov D., Chordia T., Jostova G., and Philipov A., 2007, Momentum and Credit Rating, The Journal of Finance 5, Baek, Jung, Kwon, and Moon, 2005, Transfer Entropy Analysis of the Stock Market, ArXiv Physics e-prints. Balla and Shivakumar, 2005, Earnings quality in UK private firms: comparative loss recognition timeliness, Journal of Accounting and Economics 39, Bauer, Cox, Caveness, Downs, Thornhill, 2007, Finding the Direction of Disturbance Propagation in a Chemical Process Using Transfer Entropy, IEEE Transactions on Control Systems Technology, Bernothn and Wolff, 2008, Fool the markets? Creative accounting, fiscal transparency and sovereign risk premia, Scottish Journal of Political Economy, Birz, 2017, Stale Economic News, Media and the Stock Market, Journal of Economic Psychology, Accepted Manuscript. Bleck and Liu, 2007, Market Transparency and the Accounting Regime, Journal of Accounting Research, Bhattacharya, Daouk and Welker, 2003, The World Price of Earnings Opacity, The Accounting Review 78, Bushman, Piotroski and Smith, 2004, What Determines Corporate Transparency, Journal of Accounting Research 42, Celiker U., Kayacetin N., Kumar R., Sonaer G., 2016, Cash flow news, discount rate news, and momentum, Journal of Banking and Finance 72, Chan, Jegadeesh and Lakonishok, 1996, Momentum Portfolios, The Journal of Finance 51, Chang and Chan, 2008, Financial analysts stock recommendation revisions and stock price changes, Applied Financial Economics 18, Chan W.S., 2003, Stock price reaction to news and no-news: drift and reversal after headlines, Journal of Financial Economics 70,

14 Chen and Zhao, 2012, Informed trading, information uncertainty, and price momentum, Journal of Banking & Finance 36, Cheema and Nartea, 2014, Momentum returns and information uncertainty: Evidence from China, Pacific-Basin Finance Journal, Cheema and Nartea, 2017, Momentum returns, market states, and market dynamics: Is China different? International Review of Economics and Finance, Cooper, Gutierrez Jr. and Hameed, 2004, Market States and Momentum, The Journal of Finance 59, Da, Gurun, Walachia, 2014, Frog in the Pan: Continuous Information and Momentum, The Review of Financial Studies, Dasgupta, Gan and Gao, 2010, Transparency, Price Informativeness, and Stock Return Synchronicity: Theory and Evidence, The Journal of Financial and Quantitative Analysis, Dechow, Sloan and Sweeney, 1995, Detecting Earnings Management, The Accounting Review 70, Fan J., Gillan S., and Yu X., 2013, Property rights, R&D spillovers, and corporate accounting transparency in China, Emerging Markets Review 15, Fang and Peress,2009, Media Coverage and the Cross-section of Stock Returns, The Journal of Finance 4, Fuertes A., Miffre J., Rallis G., Tactical allocation in commodity futures markets: Combining momentum and term structure signals, Journal of Banking & Finance 34, Griffin, Ji and Martin, 2003, Momentum Investing and Business Cycle Risk: Evidence from Pole to Pole, The Journal of Finance 58, Grinblatt M., Han B., 2005, Prospect theory, mental accounting, and momentum, Journal of Financial Economics 78, Goyal A., and Wahal S., 2005, Is Momentum an Echo?, Journal of Financial and Quantitative Analysis, 50, Hillert, Jacobs and Müller, 2014, Media Makes Momentum, The Review of Financial Studies 27, Hong, Lim and. Stein, 2000, Bad News Travels Slowly: Size, Analyst Coverage, and the Profitability of Momentum Portfolios, The Journal of Finance 1, Hvidkjaer S., 2006, A Trade-Based Analysis of Momentum, The Review of Financial Studies 19, Jegadeesh N. and Titman S., 1993, Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency, The Journal of Finance 48,

15 Jiang, Lee, Zhang, 2005, Information Uncertainty and Expected Returns, Review of Accounting Studies, John McInnis, 2010, Earnings Smoothness, Average Returns, and Implied Cost of Equity Capital, The Accounting Review 85, Johnson T., 2002, Rational Momentum Effects, The Journal of Finance 57, Kang J., Liu M., and Ni S., 2002, Contrarian and momentum portfolios in the China stock market: , Pacific-Basin Finance Journal 10, Kim, Yu and Zhang, 2016, Can media exposure improve stock price efficiency in China and why? China Journal of Accounting Research, Korajczyk and Sadka, 2004, Are Momentum Profits Robust to Trading Costs, The Journal of Finance 3, Lee and Swaminathan, 2000, Price Momentum and Trading Volume, The Journal of Finance 5, Lesmond, Schillb, Zhou, 2002, The illusory nature of momentum profits, Journal of Financial Economics 71, Li, Qiu, Wu, 2010, Momentum and Seasonality in Chinese Stock Markets, Journal of Money, Investment and Banking, Lin H.W., Hung M.W., Huang J.B., 2018, Artificial Momentum, Native Contrarian, and Transparency in China, Computational Economics, Accepted Manuscript. Liao, Liu and Wang, 2011, Information Discovery in Share Lockups: Evidence from the Split-Share Structure Reform in China, Financial Management, Menkhoff L., Sarno L., Schmeling M., Schrimpf A., 2012, Currency momentum portfolios, Journal of Financial Economics 106, Novy-Marx R., 2012, Is momentum really momentum?, Journal of Financial Economics 103, Rouwenhorst, 1998, International Momentum Portfolios, The Journal of Finance 1, Savor P. G., 2012, Stock returns after major price shocks: The impact of information, Journal of Financial Economics 106, Scott J., Stumpp M., and Xu P., 2003, News, Not Trading Volume, Builds Momentum, Financial Analysts Journal 59, Solomon D., Soltes E., Sosyura D., 2014, Winners in the spot light: Media coverage of fund holdings as a driver of flows, Journal of Financial Economics113, Verardo M., 2009, Heterogeneous Beliefs and Momentum Profits, Journal of Financial and Quantitative Analysis 44, Wang, 2004, Relative strength portfolios in China s stock market: , Pacific-Basin Finance Journal,

16 Wang and Chin, 2004, Profitability of return and volume-based investment portfolios in China s stock market, Pacific-Basin Finance Journal, Wu, 2011, Momentum trading, mean reversal and overreaction in Chinese stock market, Review of Quantitative Finance and Accounting, Yang X., and Luo Y., 2014, Rumor Clarification and Stock Returns: Do Bull Markets Behave Differently from Bear Markets?, Emerging Markets Finance & Trade 50,