کدام معاملات قیمتها را در بازارهای در حال ظهور تکان می دهد؟ شواهدی از بازار سهام چین

This paper extends Barclay and Warner's [Barclay, M.J. and J.B. Warner (1993), ‘Stealth trading and volatility: which trades move prices?’, Journal of Financial Economics, vol. 34, pp. 281–306.] original work on stealth trading by analysing which trades move price for the emerging Chinese stock market. A large block trade/manipulation hypothesis is proposed in addition to the stealth and public information hypotheses examined by Barclay and Warner. Using high-frequency data the results show that while medium and large-size trades are associated with disproportionately large, overall, cumulative stock price changes, it is the large-size trades (in terms of the number of transactions) which have the largest effect on cumulative price increases. Thus, while there is some support for stealth trading in the Chinese market, there are other effects in operation such as large block trades/price manipulation.

This paper extends Barclay and Warner's (1993) work on stealth trading by both analysing which trades move price in the emerging Chinese stock market and by proposing/testing a large block trade/manipulation hypothesis in addition to the stealth and public information hypotheses considered by Barclay and Warner. The central prediction of the stealth-trading hypothesis is that privately informed traders use medium sized trades so as not to reveal their information (as would be the case if they used large trades) and for a sample of 105 NYSE firms that were tender-offer targets between 1981 and 1984, Barclay and Warner found most of the cumulative stock-price change is due to medium-size trades. Similarly, using audit trail data for a sample of NYSE firms, Chakravarty (2001) found that medium-size trades are associated with a disproportionately large cumulative stock price change relative to their proportion of all trades and volume. In addition, they found that the source of the disproportionately large cumulative price impact of medium-size trades is trades initiated by institutions. Finally, Anand and Chakravarty (2003) found that informed traders tend to operate in medium sized trades of high leverage and high volume options. The extension of the analysis of stealth trading from a primarily US context to China is of interest for a number of reasons. First, China is an emerging market with a developing regulatory framework and this allows us to examine whether traders adopt stealth trading in a less developed regulatory regime. Second, it is unclear whether there would be the same need for privately informed investors to adopt stealth trading given the relative youth of the Chinese stock markets and the possible lack of sophistication of the general investing public. Third, given these first two factors there may be scope for traders to gain by other trading strategies such as price manipulation and this possibility is examined here. Finally, China is an emerging global power which has already and will continue to attract businesses and investors from the rest of the world. These organisations and individuals will want to understand the efficiency of the Chinese stock market especially given the concerns expressed by a number of eminent individuals in China. At the National People's Congress (NCP) in spring 2000, Premier Zhu Rongji remarked that China's stock markets had developed quickly, achieved much but were still not well regulated with concerns over rampant speculation, poor-quality listed firms, defective regulation and widespread corruption. In 2001, Professor Wu Jinglian commented that ‘China's stock market is no better than a casino. At least in a casino there are rules.’ (Wu, 2001). The current results show that while stealth trading and large block trades/price manipulation hypotheses have explanatory power for overall price changes, it is large block trades/price manipulation which best explains price increases. Therefore, these results add to the existing literature on stealth trading by showing that while it still exists in the context of the emerging Chinese market with all of its associated characteristics, there are other forces at work in this emerging market. The results may have implications for the regulatory oversight of stock trading in China and for research on stealth trading. In terms of the latter, further work could look at other emerging markets and different regulatory regimes. The structure of the paper is as follows. Section 2 presents the various hypotheses, Section 3 discusses the data and methodology, while Section 4 presents the results and Section 5 offers conclusions and suggestions for further research.

This article has extended Barclay and Warner's (1993) work on stealth trading to the emerging Chinese stock market and to the consideration of a large block trades/price manipulation hypothesis. While some support is found for stealth trading in the analysis of overall price changes and price decreases, the results point towards large size trades having the greatest disproportionate effect on price changes and this is especially the case for price increases. However, caution needs to be shown in arguing that these results are a pure consequence of price manipulation. It may simply be the case that informed traders do not need to adopt stealth trading because the other investors are relatively unsophisticated. If this is indeed the case, large trades could reflect both informed traders and manipulated traders. Moreover, if manipulators are known to use large trades, then informed traders should adopt a similar strategy because other investors will not be able to separate informed from manipulated trades.2 We believe these results add to the current literature on stealth trading and suggest the need for further work. In particular, if support is found for a ‘large trade’ effect, then additional analysis could be usefully conducted to identify the precise causes of such an effect; more specifically, is it an informed and/or manipulated trade effect? The separation of informed and manipulated trades might be achieved via two routes. First, an analysis of the time a position is held might be illuminating because informed trades, in general, might be expected to be held longer than manipulated trades. Second, a comparison of large price changes conditional/non-conditional on news announcements might also be helpful. Though, it has to be accepted that any results will be noisy as theoretical predictions of the behaviour of informed traders and price manipulators around news announcements is not well defined.