Fat‐Finger Trade and Market Quality: The First Evidence From China

• Published date: 01 October 2016
• Author: Ming Gao,Yu‐Jane Liu,Weili Wu
• DOI: http://doi.org/10.1002/fut.21771
• Date: 01 October 2016

Fat-Finger Trade and Market Quality:

The First Evidence From China

Ming Gao*, Yu-Jane Liu and Weili Wu

More trading is algorithmic or computer generated, and in markets where it is allowed, high

frequency. However, what happens when there is an algorithmic trading error? This study

attempts to answer that question by examining the August 16,2013, fat-finger trade in Chinese

equity and equity futures markets. We find that both markets were excessively volatile, illiquid,

and positively skewed. Moreover, we document that index returns are predictable for a short

time, indicating that the fat-finger event induced an inefficient market. Our results highlight

the importance of market surveillance and regulation to lessen the damage of future fat-finger

events. © 2016 Wiley Periodicals, Inc. Jrl Fut Mark 36:1014–1025, 2016

1. INTRODUCTION

On Aug 16, 2013, there was an unexpected large increase in the CSI 300 index and its trading

volume within three minutes of an erroneous computer program placing trades by China

Everbright Securities, the country’sfifth-largest brokerage firm by market value. This fat-

finger event was a new experience for Chinese financial markets and induced investor

responses that were rapid and profound.

This paperis one of the first to examine the external shocks caused by this fat-finger event.

The objective of this paper is to analyze the adverse impact of an algorithmic trading error,

given that more trading is nowalgorithmic or computer generated. We examine the effectthe

shock had on market qualityand explore the dramatic change in intradaymarket behavior due

to Everbright’s fat-finger trade in order to understand the intraday price distribution and

correlation dynamics under market stress in a market dominated by retail investors. The

Everbright Securities fat-finger episode suggests that algo trading errors can cause price

shocks even in markets where high-frequency trading is not allowed, such as in China.

Other events similar to the Ev erbright Securities fat-finger event have transpired. On

August 1, 2012, a computer at Kni ght Capital Group, one of the maj or market-making

firms in the U.S., executed a serie s of automatic orders that were ap parently buying high

and selling low, the opposite o f a competent trading strategy. Knight’s problem was cau sed

by incorrectly written code in a new computer system. Appare ntly, the company did not

have its old system running at t he same time, which is standard protocol when introducin g

Ming Gao is at Peking University, Beijing, China. Yu-Jane Liu is at Guanghua School of Management, Peking

University, Beijing, China. Weili Wu is at Central University of Finance and Economics, Beijing, China. The

authors gratefully acknowledge the helpful comments and suggestions made by the editor and referees.

JEL Classification: G14

*Correspondence author, Peking University, 5 Yiheyuan Road, Haidian District, Beijing, 100871, China. Tel: þ86

10 62753626, Fax: þ86 10 62753776, e-mail: gao@pku.edu.cn

Received July 2015; Accepted November 2015

The Journal of Futures Markets, Vol. 36, No. 10, 1014–1025 (2016)

© 2016 Wiley Periodicals, Inc.

Published online 4 February 2016 in Wiley Online Library (wileyonlinelibrary.com).

DOI: 10.1002/fut.21771