High‐Frequency Positive Feedback Trading and Market Quality: Evidence from China's Stock Market

• Date: 01 December 2017
• DOI: http://doi.org/10.1111/irfi.12116
• Published date: 01 December 2017
• Author: Die Wan,Xiaoguang Yang

High-Frequency Positive Feedback

Trading and Market Quality:

Evidence from China’s Stock Market*

DIE WAN

†

AND XIAOGUANG YANG

‡

†

School of Finance, Zhejiang Gongshang University, Hangzhou, China and

‡

Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing,

China

ABSTRACT

This paper managed to measure the positive feedback trading intensity and its

asymmetry with high-frequency transaction data of China’s individual stocks.

The intraday positive feedback trading is found to be heterogeneous, and

buying-winners effect is significantly stronger than selling-losers effect. In gen-

eral, the high-frequency asymmetric positive feedback trading’s impact on

market quality is mixed: The intraday positive feedback trades contribute to

a liquid and active-trading market but at the same time slow down the price

discovery process and reduce the price efficiency.

JEL Codes: G02; G14; G15

I. INTRODUCTION

Positive feedback trading relates to transaction on the basis of historical prices

and involves buying stocks when the market is improving and selling stocks

when the market is declining. In the preliminary model of DeLong et al.

(1990b), positive feedback traders are assumed to be irrational traders, and their

trading destabilizes the market. Pritsker (1997)’s model further shows that posi-

tive feedback trading may lead to illiquid market and harm the market quality.

But the direct evidence of positive feedback trading reducing price efficiency is

actually limited. Empirical studies mostly show indirect evidence of positive

feedback trading’s contribution to inefficient market. Sentana and Wadhwani

(1992) assume the existence of positive feedback traders and show that their trad-

ing leads to negatively autocorrelatedreturns. Jegadeesh andTitman (1993) suggest

that the profitabilityof quarterly or yearly momentum investing is just evidence of

* This research was supported by the National Natural Science Funds with Grant numbers

(71431008, 71532013, and 71501170), Zhejiang Provincial National Science Funds (No.

LQ16G010001), and Zhejiang Provincial Key Research Base for Humanities and Social Science Re-

search (Applied Economics in Zhejiang Gongshang University). The authors are grateful for the in-

sightful comments and suggestions provided by the editor and two referees, which have helped

improve the paper significantly. All errors are our own.

© 2017 International Review of Finance Ltd. 2017

International Review of Finance, 2017

DOI: 10.1111/irfi.12116

International Review of Finance, 17:4, 2017: pp. 493–523

DOI: 10.1111/irfi .12116

© 2017 International Review of Finance Ltd. 2017

market inefficiency. However, Johnson (2002) provides a reverse argument that

challenges the irrationality of positive feedback trading. He builds up a model to

show that rationalmomentum investing may be possiblethrough various expected

dividendgrowth rates, and thus positive feedbacktrading may also help to improve

the market efficiency. The research resultsof the relation between positivefeedback

trading and market quality are not consistent and need to be further explored.

This paper seeks to investigate the impact of high-frequency positive feedback

trading on market quality in China’s stock market. The choice for China’s stock

market is based on the following considerations. First, China’s stock market in

the market value is the largest emerging market and the second largest stock mar-

ket in the world, and China’s stock market has the biggest investor population in

the world. So China’s stock market itself is valuable for studying what happens be-

tween positive feedback trading and market quality in China and what the differ-

ence is between China’s story and the stories in other markets. Second, the retail

trading is quite intensive in China,

1

and if retail traders herd more as suggested in

Barber et al. (2009), the positive feedback trading intensity may be salient and

time-varying. Moreover, the short-selling constraints on retail traders may inten-

sify positive feedback trading and amplify the variations in the levels of asymme-

try between buying-winners and selling-losers effects.

2

Griffin et al. (2007) have

actually shown that the positive relation between turnover and past return is

stronger in markets with short-sale constraints and high volatility. Because of

these features discussed in the preceding texts, China’s stock market provides a

good instance for studying the impact of positive feedback trading. Third, high-

frequency data help to closely examine the relation between investors trading be-

haviors and their market impact. Previous studies either rely on a long-term data

to estimate the parameter of feedback intensity (Sentana and Wadhwani 1992) or

use daily time series to estimate regression coefficients (Griffin et al. 2003; Wan

et al. 2016). These results from low-frequency data provide an average estimation

of feedback trading intensity on the whole market or the entire sample period,

while our measures based on high-frequency data could take account of the vari-

ations of feedback intensity across time and firms to reach a more precise study.

In fact, Wan et al. (2016) find that positive feedback trading in China’s indi-

vidual stock is more intensive when stock price goes up, which is just reverse to

the findings in index data (Sentana and Wadhwani 1992; Koutmos 1997;

Koutmos and Saidi 2001; Hou and Li 2014). Wan et al. (2016) relate the asymme-

try to the herding behaviors of retail traders. They show that more retail trading

in individual stock leads to more buying-winners effect, while more rational trad-

ing like stop-loss strategies contributes to more selling-losers effect. So far, no lit-

erature answers the questions whether there is any asymmetry of positive

1 According to the 2012 annual statistics of China Securities Depository and Clearing Corpora-

tion Limited, for more than 97.4% of the investors’accounts, the market value in each ac-

count is less than 500,000 Yuan. And more than 85% of the market trading volumes are

produced by these accounts.

2 The capital limit for short selling in China’s stock market is 500,000 Yuan, much higher than

the average capital in an ordinary investor’s account.

International Review of Finance

© 2017 International Review of Finance Ltd. 20172

International Review of Finance

494 © 2017 International Review of Finance Ltd. 2017