Endogenous informed trading in the presence of trading costs: Theory and evidence

The basic premise of the model we propose is that market frictions (trading costs) force traders with market-wide information to strategically choose which securities to trade in. We study the effect of recognizing trading costs on the choices of informed traders and the resulting statistical properties of security prices. Specifically, we show that (1) stocks with intermediate β's have the least informative prices, even though they are traded by the greatest number of informed traders; (2) for high β securities, the contemporaneous correlation of prices is close to the correlation in fundamental values; (3) a security with a higher β, higher volume of liquidity trading and lower idiosyncratic variance is more likely to lead another security. With market capitalization as a proxy for the level of liquidity trading, these specific predictions of the model on the lead–lag relationship are also shown to be strongly supported by the data.     

The determinants of quantile autocorrelations: Evidence from the UK

We empirically analyse the cross-sectional determinants of stock return autocorrelations in the UK in different quantiles of conditional return distributions. Autocorrelations in low quantiles are predominantly positive, whereas those in the remaining quantiles are negative. Autocorrelations in different quantiles depend on different sets of firm and trading characteristics: when returns are normal or high, prices react quickly to information, are driven by positive feedback traders, instantaneous news arrivals, and overshoot, trades are predominantly motivated by hedging/liquidity needs, and measured autocorrelations can be biassed by the bid–ask bounce effect and nonsynchronous trading. However, when returns are unusually low, prices are driven by information arriving sequentially and react sluggishly to it, and are influenced by trading on private information and/or negative feedback traders.     

The January Effect and Aggregate Insider Trading

This study investigates the seasonal pattern of aggregate insider trading to help distinguish between two competing explanations for the seasonal pattern of security returns. The first potential explanation examined is that the January effect arises from predictable changes in turn-of-the-year demand for securities. The second potential explanation examined is that the January effect represents compensation for the higher risk of trading against informed traders at the turn of the year.     

Multimarket trading and market liquidity

When a security trades at multiple locations simultaneously,an informed trader has several avenues in which to exploit hisprivate information. The greater the proportion of liquiditytrading by 'large' traders who can split their trades acrossmarkets, the larger is the correlation between volume in differentmarkets and the smaller is the informativeness of prices. Weshow that one of the markets emerges as the dominant locationfor trading in that security. When informed traders can usetheir information for more than one trading period, the timelyrelease of price information by market informed traders expectto make subsequently at other locations. Markets makers, competingto offer the lowest cost of trading at their location, consequentlydeter informal trading by voluntarily making the price informationpublic and by 'cracking down' on insider trading.     

Trading frequency and volatility clustering

Volatility clustering, with autocorrelations of the hyperbolic decay rate, is unquestionably one of the most important stylized facts of financial time series. This paper presents a market microstructure model that is able to generate volatility clustering with hyperbolically decaying autocorrelations via traders with multiple trading frequencies, using Bayesian information updates in an incomplete market. The model illustrates that signal extraction, which is induced by multiple trading frequencies, can increase the persistence of the volatility of returns. Furthermore, we show that the volatility of the underlying time series of returns varies greatly with the number of traders in the market.     

卖空者与内幕交易——来自中国证券市场的证据

本文研究上市公司内部人减持、年报、诉讼、分析师评级、停复牌以及高送转等重大公告前后卖空交易行为的变化,系统考察卖空者是否参与内幕交易以及何种因素影响卖空者参与内幕交易,发现卖空率较高的股票具有较低的未来收益,表明卖空者拥有信息优势,属知情交易者;卖空者拥有非常精确的择时交易能力,在重大利空公告前显著增加卖空量,而在利好公告前则显著减少卖空头寸,表明卖空者作为知情交易者的信息优势源自内幕消息;公司内、外部投资者的信息不对称程度越低或公司所在地的法治水平越高,卖空者参与内幕交易的行为就越少。因此,监管机构应密切关注公司重大消息发布前后卖空量的异常变动,同时,完善信息披露规则、健全证券分析师制度并强化法律法规的执行力度,才能有效防范卖空者参与内幕交易。     

Exploitable patterns in retirement annuity returns: evidence from TIAA/CREF

Evidence suggests that predictabilities in asset class returns exist but transactions costs prevent exploiting them using individual securities. Extant research also shows that these relationships may by exploitable through the trading of mutual funds but fails to examine whether this relationship exists within an individual fund family. This paper finds that TIAA/CREF retirement annuities exhibit predictable elements that could be exploited by informed traders. The proposed trading strategy dominates a buy-and-hold strategy by producing higher raw and risk-adjusted returns. Additionally, the risk is greatly reduced.     

An empirical analysis of the incentives to engage in costly information acquisition: The case of risk arbitrage

In order for security prices to be informationally efficient, incentives must exist for traders to engage in costly information acquisition. This paper provides empirical evidence on this proposition. We observe that risk arbitrageurs (i.e., market participants who trade in securities of firms that are involved in mergers, tender offers, and voluntary liquidations) are able to generate private information regarding the success or failure of corporate reorganizations. Moreover, risk arbitrageurs earn substantial returns on their trading activities. These results suggest that security prices are sufficiently noisy to create incentives for costly information acquisition.     

Are margin traders informed?

The literature offers contradictory views on the informativeness of margin trading using various measures of information content. Utilising data from a Chinese margin‐trading pilot programme initiated in 2010, this paper investigates whether margin traders are informed by directly examining the return predictability of margin‐trading activity. We find that margin‐trading activities cannot positively predict future stock returns. Moreover, we explore some non‐informational trading strategies used by margin traders, e.g., positive‐feedback strategies and moving‐average trading rules. These results suggest that margin traders are noise traders rather than informed traders, and margin trading conveys no new firm‐specific information.     

Noise trading and prime and score premiums

This paper documents that a common element drives the time-series variation of the premium pricing of Primes and Scores. I argue that this common element is noise trading. The noise trading model of Delong, Shleifer, Summers and Waldmann (1990) predicts that returns on assets that are predominantly traded by noise traders will be correlated, since the misperceptions of noise traders are cross-sectionally correlated. Consistent with the noise trading hypothesis, changes in the average premium of Primes and Scores, which are predominantly traded by individual investors, are correlated with both changes in average discounts of closed-end funds and small firms returns. These empirical facts provide additional evidence that noise traders can affect security prices.     

The information content of trading halts

We investigate the impact of trading halts of NYSE-listed stocks on informationally related securities that continue to trade during the period of the halt. Informational relationships are established for companies in the same four-digit SIC industry based on the correlation of returns, volume, volatility, and the adverse selection components of spreads. We find a significant liquidity impact on informationally related securities with spreads and price impact of trades having substantial increases. However, we also find that quoted depths, the number of trades, and trade volume significantly increase. Our results are consistent with the trading halt model of Spiegel and Subrahmanyam [2000. Asymmetric information and news disclosure rules. Journal of Financial Intermediation 9, 363–403] and with the informed trading model of Tookes [2008. Information, trading, and product market interactions: cross-sectional implications of informed trading. Journal of Finance 63, 379–413]. In addition, our results indicate that there is a common liquidity response of informationally related securities to firm-specific trading halts.     

Informed Futures Trading and Price Discovery: Evidence from Taiwan Futures and Stock Markets

This article focuses on the information effects between the futures market and its spot market. Intraday data are used to investigate the lead-lag relationships between the returns and trading activity of Taiwan stock index futures and the spot returns. We focus on the transmission direction and the sources of information. Consistent with most previous studies, our results show that other than the contemporaneous relationship predicted by carry-cost theory and efficient market theory, futures returns significantly lead spot returns, which implies that informed trades may occur in the futures market. Using private transaction information, net open buy, as a proxy for futures trading activity and distinguishing different types of futures traders, we find that foreign institutional traders are the major source of informed trades because their trading has predictive power for future movements in both spot and futures prices. Traders in other categories are information laggards.     

Trading frequency and the efficiency of price discovery in a non-dealer market

The increasing popularity of non-dealer security markets that offer automated, computer-based, continuous trading reflects a presumption that institutionally-set trading sessions are economically obsolete. This theoretical paper investigates the effect of the trading frequency, a key feature of the trading mechanism, on the efficiency of price discovery in a non-dealer market. By tracing the market pricing error to the correlation structures of arriving information and pricing errors of individual traders, the effect of diverging expectations on error-based and overall return volatility is isolated. The analysis reveals that, due to a portfolio effect, an increase in the trading time interval has contradictory effects on the portion of return volatility stemming from pricing errors. The greater accumulation of information increases error-based return volatility, but the greater volume and number of traders per session have the opposite effect. The net effect on overall return volatility can go either way. It is found that the return volatility of heavily traded securities is likely to be minimized under continuous trading, but that of thinly traded securities may be minimized under discrete trading at moderate time intervals. The latter is more likely to occur the greater is the divergence of expectations among traders. These findings challenge the presumption that automated continuous trading in a non-dealer market is more efficient than discrete trading for all securities, regardless of trading volume. The findings are applicable to all economies, but have special importance for developing countries where typically a single market is dominated by small issues and a low volume of trade. As a by-product of the analysis, it is shown how to correct the biased estimate of inter-session price volatility when observations are less frequent than the trading sessions themselves.     

Strategic Trading When Agents Forecast the Forecasts of Others

We analyze a multi-period model of trading with differentially informed traders, liquidity traders, and a market maker. Each informed trader's initial information is a noisy estimate of the long-term value of the asset, and the different signals received by informed traders can have a variety of correlation structures. With this setup, informed traders not only compete with each other for trading profits, they also learn about other traders' signals from the observed order flow. Our work suggests that the initial correlation among the informed traders' signals has a significant effect on the informed traders' profits and the informativeness of prices.     

Trade Size Preference of Informed Traders in Indian Equity Markets

This study utilized high frequency transactions data to analyze the trade size preference of informed traders in Indian equity markets. It is observed that informed traders at an aggregate level adopt stealth trading strategy, wherein they prefer medium sized trades over large sized trades in order to camouflage their private information. However, the stealth trading behavior varies across stocks, wherein informed traders prefer more large sized trades on firms that are part of an index compared to non-index firms. Trading behavior also varies across other market conditions. It has been noted that informed traders prefer large sized trades during periods of high market thickness, negative returns, and low volatility. This study also provides a rationale for such varied behavior of informed traders.     

Trade Disclosure,Information Learning and Securities Market Performance

In this paper we examine the effect of information disclosure on securities market performance when liquidity traders are able to acquire information about inside trading. We show that the bid-ask spread increases with the liquidity trader's learning efficiency, which is greater when trade information is disclosed. The bid-ask spread is always higher when trade information is not disclosed. However, the discrepancy between the bid-ask spreads with and without information disclosure narrows when the learning efficiency increases. We also show that the gains of the informed traders in a market without trade information disclosure are reduced in the presence of the liquidity trader's learning. Nevertheless, liquidity traders do not necessarily benefit from increased transparency. In particular, liquidity traders may face higher trading costs.     

Informed and uninformed trading on the Australian dollar

Using a high-frequency data set of the spot Australian/US dollar, this study examines the distribution of quotes, spreads, and returns across the trading day. By identifying the direction of trade and the subsequent quote returns from contributing banks, the segmented nature of the market into market-makers and informed and uninformed traders is investigated. The results suggest that the economic gain possible from private information is maximised over 2 to 5 quotes and is rapidly eroded by 20 quotes (about 2 min later during busy trading times) as other new information enters the market. Also, the analysis is revealing of discontinuities in trading and the volatility of pricing across the trading day.     

Odd lot trading and earnings announcements

This study investigates odd lot trading, both trades and orders, around quarterly earnings announcements to determine whether odd lot traders are informed regarding the information contained in earnings announcements. We find pre-announcement odd lot order imbalances are not positively correlated with post-announcement returns and odd lot traders do not earn excess returns. Portfolios long stocks highly bought by odd lot traders in the pre-announcement period and short stocks highly sold by odd lot traders do not outperform the market. We conclude that odd lot traders are not in possession of earnings announcement information prior to its release to the public.     

Trading Volume and Cross-Autocorrelations in Stock Returns

This paper finds that trading volume is a significant determinant of the lead-lag patterns observed in stock returns. Daily and weekly returns on high volume portfolios lead returns on low volume portfolios, controlling for firm size. Nonsynchronous trading or low volume portfolio autocorrelations cannot explain these findings. These patterns arise because returns on low volume portfolios respond more slowly to information in market returns. The speed of adjustment of individual stocks confirms these findings. Overall, the results indicate that differential speed of adjustment to information is a significant source of the cross-autocorrelation patterns in short-horizon stock returns.     

Infrequent Rebalancing,Return Autocorrelation,and Seasonality

A model of infrequent rebalancing can explain specific predictability patterns in the time series and cross‐section of stock returns. First, infrequent rebalancing produces return autocorrelations that are consistent with empirical evidence from intraday returns and new evidence from daily returns. Autocorrelations can switch sign and become positive at the rebalancing horizon. Second, the cross‐sectional variance in expected returns is larger when more traders rebalance. This effect generates seasonality in the cross‐section of stock returns, which can help explain available empirical evidence.