Exercise to Lose Money? Irrational Exercise Behavior from the Chinese Warrants Market

• Author: Weiqiang Zhang,Zhisheng Li,Li Liao,Ning Zhu
• DOI: http://doi.org/10.1002/fut.21608
• Date: 01 May 2014
• Published date: 01 May 2014

EXERCISE TO LOSE MONEY?

IRRATIONAL EXERCISE BEHAVIOR

FROM THE CHINESE WARRANTS MARKET

LI LIAO, ZHISHENG LI, WEIQIANG ZHANG and NING ZHU*

Using a market‐level exercise data set and an individual‐level trading data set between

August 2006 and June 2009, this study examinesthe incidence of two types of irrational exercise

behavior in the Chinese warrants market. We find that 121.64 million shares of warrants (0.64%

of all warrants) were either exercised with an immediateloss or failed to be exercised, resulting in

foregone risk‐free profits. These irrational exercises caused warrant holders to lose over 717.79

million Yuan. Some of the irrational behavior can be attributed to “entertainment seeking”and

the “Tþ1”rule practiced in the Chinese security market, but the majority is attributed to

warrant holders’ignorance and/or negligence of warrant mechanics. Our findings provide

additional field evidence of clearly irrational exercise behavior in a derivatives market. We also

find that investor education, information and guidance provision can mitigate the incidence of

irrational exercise behavior significantly. © 2013 Wiley Periodicals, Inc. Jrl Fut Mark

34:399–419, 2014

1. INTRODUCTION

Evidence of behavioral mistakes in the financial markets is increasing. Given that increasing

studies (Barber, Lee, Liu, & Odean, 2009; Barber, Odean, & Zhu, 2009; Hvidkjaer, 2008;

Kaniel,Saar, & Titman, 2008;Kumar & Lee, 2006) showthat such irrationalbehavioral patterns

by investors can have substantial impact on price formation and liquidity in the securities

markets, more needs to be known as to why investors make some seemingly simple mistakes.

One challenge in answering such a q uestion is that prior studies rely heavily on

specific assumptions about invest or utility (Campbell, 2006; Dhar & Zhu, 20 06; Grinblatt

& Keloharju, 2001; Heath, Huddar t, & Lang, 1999; Lee, Liu, & Zhu, 2008; Odean, 19 98),

Li Liao is a Professor at PBC School of Finance at Tsinghua University, Beijing, People’s Republic of China.

Zhisheng Li is a Professor at the School of Finance at Zhongnan University of Economics and Law, Wuhan,

People’s Republic of China. Weiqiang Zhang is a Postdoctoral Researcher in School of Economics and

Management at Tsinghua University, Beijing, People’s Republic of China. Ning Zhu is a Professor at Shanghai

Advanced Institute of Finance (SAIF), Shanghai Jiaotong University, Shanghai, People’s Republic of China,

and UC Davis, Davis, California. We are very grateful to the anonymous brokerage firm for making available

the data used in the study and answering questions related to the data. The authors benefited from comments

from Charles Cao, Neil Pearson, seminar participants at SAIF, an anonymous referee, and Bob Webb (the

editor). We acknowledge financial support from the Tsinghua‐Citi Financial Research and Education

Program. Liao, Zhang, and Zhu acknowledge Project 71232003 supported by NSFC. Li acknowledges project

71271214 and 70801063 supported by NSFC. Zhu acknowledges Project 70972009 and 70932002 supported

by NSFC. All remaining errors are ours only.

*Correspondence author, Shanghai Jiaotong University, Shanghai, People’s Republic of China, and UC Davis, Davis,

CA. Tel: þ86‐21‐6293 2280, Fax: þ86‐21‐6293 2310, e‐mail: nzhu@ucdavis.edu

Received October 2011; Accepted January 2013

The Journal of Futures Markets, Vol. 34, No. 5, 399–419 (2014)

© 2013 Wiley Periodicals, Inc.

Published online 19 March 2013 in Wiley Online Library (wileyonlinelibrary.com).

DOI: 10.1002/fut.21608

whereas derivatives markets p rovide a unique opportunity to study inv estor behavior

because the valuation of many der ivative securities can be calculated pre cisely without

invoking particular assumptions ab out expectation or investor utility function. Fol lowing

such logic, prior studies indee d document that investors fail to properly exercise American

options before expiration in the U.S. options market (Diz & Finuc ane, 1993; Overdahl,

1988; Pool, Stoll, & Whaley, 2008). Al though some studies point out that market fricti on

and institutional considerati ons are responsible for the irrational exerci se behavior

(Dawson, 1996; Pool et al., 2008), Po teshman and Serbin (2003) show that part of the

irrational exercise behavior can be a ttributed precisely to irrational behaviora l decisions.

Building upon the previous studies on early option exercise behavior, the current study

looks at a different aspect of optimal option exercise behavior in the context of the Chinese

warrants market, where investors face a relatively simple decision upon warrant expiration:

whether or not to exercise the warrants that they hold.

1

Because the strike price and the market price of the underlying stock are easy to identify,

it seems straightforward to decide whether to exercise a particular warrant. However, we

indeed find that Chinese investors frequently fail to exercise valuable warrants upon

expiration and sometimes exercise out‐of‐the‐money warrants that cause them to lose

money. Among all 39 warrants, we find irrational exercise behavior of some kind for 35

warrants (89.74% of the sample). In terms of the total number of expiring warrant shares,

121.64 million out of 19,002.12 million shares were identified as irrational exercise (0.64%).

Our calculations reveal that investors have lost over 700 million Yuan (over 100 million USD)

by failing to properly exercise the warrant contracts that they hold.

By investigating the incidenc e of irrational exercise behavior for differe nt types of

warrant contracts, we gain so me insights for reasons behind the app arent irrational

exercise behavior. First, we find that the “Tþ1”del ivery mechanism in the Chinese stoc k

market is partly responsibl e for our findings. Because investor s would have to hold the

shares that they receive from w arrant exercise for at least 1 day befor e they can liquidate

such stocks in the stock market for call warrants and pur chase the underlying shares at

least one trading day prior to exe rcise for put warrants, they face poten tial risks from the

next day price movement. Once we control for the 10% price movement limit in the next

day of trading (the maximum p otential risk one may face when decid ing whether to

exercise the warrant), the frac tion of irrational exercise decreases to 0.53 % of all warrant

shares (from 0.64% when not ac counting for the “Tþ1”trading rule), yet still remains

prevalent.

Secondly,we find that part of the irrational exercise behavior can be attributedto investors’

lack of knowledge about warrant contracts and their lack of attention when warrant contracts

come to expiration. Investors are far less likely to irrationally exercise their warrants at the

Shenzhen Stock Exchange (0.10% of all warrant shares), which requires warrant issuers to

publicly disclose the valueof warrants and the optimal exercise decision upon expiration,than

those at the Shanghai Stock Exchange (0.74% of all warrant shares), which does not impose

similar requirement for information disclosure related to warrant exercise.

The current study provides two primary contributions to the extant literature. First, we

provide some novel evidence of clearly irrational trading behavior when Chinese investors

trade derivatives in the emerging derivatives market.

The Chinese warrants market creates at least three unique opportunities for studying

irrational investor behavior. First, unlike other previously documented behavioral biases in

investor trading that require certain assumptions about investors’utility functions (Dhar &

1

In general, the stock exchange or brokerage firms do not automatically exercise in‐the‐money warrants on behalf of

investors.

400 Liao et al.