How do US options traders “smirk” on China? Evidence from FXI options

• Author: Sebastian A. Gehricke,Jianhui Li,Jin E. Zhang
• Date: 01 November 2019
• Published date: 01 November 2019
• DOI: http://doi.org/10.1002/fut.22005

Received: 3 February 2019

|

Accepted: 3 February 2019

DOI: 10.1002/fut.22005

RESEARCH ARTICLE

How do US options traders “smirk”on China? Evidence

from FXI options

Jianhui Li

|

Sebastian A. Gehricke

|

Jin E. Zhang

Department of Accountancy and Finance,

Otago Business School, University of

Otago, Dunedin, New Zealand

Correspondence

Jianhui Li, Department of Accountancy

and Finance, Otago Business School,

University of Otago, Dunedin 9054, New

Zealand.

Email: liji1479@student.otago.ac.nz

Funding information

University of Otago; National Natural

Science Foundation of China, Grant/

Award Number: 71771199

Abstract

In this paper, we study the implied volatility smirk (IVS) of options written on

the FXI, the Financial Times Stock Exchange/Xinhua China 50 Index exchange‐

traded fund (ETF). Using the methodology of Zhang and Xiang (2008, Quant

Financ, 8, pp. 263–284), we document the empirical characteristics of the level,

slope, and curvature of IVS of the FXI options. We find that, on average, IVS

becomes steeper and more convex as time to maturity increases. The level and

curvature are usually positive, and the slope is negative. We provide evidence

that the information in the quantified IV factors has some predictive power for

the future monthly FXI ETF returns.

KEYWORDS

China equities, FXI options, FXI smirk, implied volatility

1

|

INTRODUCTION

This study quantifies and examines the shape and dynamics of the implied volatility (IV) of iShares China Large‐Cap

exchange‐traded fund (ETF) options, and tests the quantified IV factors as predictors of the underlying monthly FXI

ETF returns. The FXI options market has become the largest and most liquid China‐related options market. This is the

first paper concentrating on the FXI options market and documents the empirical features of the IV smirk of FXI

options. We adopt and expand the methodology developed by Zhang and Xiang (2008) to quantify the IV by fitting a

quadratic function. This results in three IV factors: the level, slope, and curvature. We further develop the constant

maturity IV factors to study the term structure and time‐series dynamics more accurately. On average, the FXI IV curve

exhibits a smirk shape, similar to that of S&P 500 options. As the maturity of FXI options increases, the IV smirk

becomes steeper and more convex. US options traders smirk on China just as they do on the United States. They prefer

to buy out‐of‐the‐money (OTM) put options to hedge the risk of market crashes. We also find that the first difference of

the third cumulants, derived from the factors, has some predictability of the future FXI returns. The empirical features

we present provide implications for the development of an FXI option pricing model and for traders to better

understand this market.

FXI option contracts are traded in the United States and have become the largest and most active options targeted on

Chinese equities available to global traders. The underlying ETF, FXI, seeks to replicate the performance of the

Financial Times Stock Exchange (FTSE) China 50 Index. In 2001, when it was first launched, the index consisted of the

25 largest‐capitalization Chinese equities that trade on the Hong Kong stock exchange. After a tremendous expansion in

volume of the Chinese equity market, on September 22, 2014, the index expanded from 25 to 50 constituents. The FXI

tracks the performance of the FTSE China 50 Index very closely, as can be observed in Figure 1. The FXI slightly

underperforms the underlying index due to the fund fees. In Table 1, we can see that the FXI ETF has a larger market

and is more liquid than other US‐traded China‐targeted ETFs, and therefore, it is the most important fund providing

exposure to Chinese equities. From reading the news on Chinese equities, it is obvious that the FXI is a reflection of the

J Futures Markets. 2019;39:1450–1470.wileyonlinelibrary.com/journal/fut1450

|

© 2019 Wiley Periodicals, Inc.

opportunities for investment in the economy of China. For example, the tariff war with America is among the factors

depressing stocks in China recently and making some traders go bearish on FXI. “One options trader is betting on

bigger losses for Beijing’s big‐cap stocks, targeting iShares China Large‐Cap ETF (FXI) put options in today’s trading,”

notes Schaeffer’s Investment Research (Venema, 2018). “Without a clear answer, a recent 6% run‐up in the iShares

China Large‐Cap exchange‐traded fund (ticker: FXI) looks like a one‐off driven by an off‐again turn in trade tensions

and the June 1 inclusion of some Chinese A‐shares in MSCI global indexes”—Wall Street Journal (Mellow, 2018).

In this paper, we first study the shape and dynamics of the IV of FXI options. We find that the FXI IV usually

exhibits a smirk shape. The overall level, which estimates the exact at‐the‐money IV (ATM IV), and slope, are usually

positive and negative, respectively, while the curvature fluctuates around zero with a positive mean. The term structure

of the level is upward sloping, while the term structures of the slope and curvature are downward sloping, on average.

We also explore the time‐series dynamics of the FXI IV curves and find that the level (ATM IV) and curvature mean‐

revert above zero while the slope is mostly negative. The level (ATM IV) factor mean‐reverts with prolonged periods of

high values (high ATM IV) during economic downturns, such as the global financial crisis (GFC). The spikes in the

long‐maturity slope and curvature factors are larger in magnitude and more frequent than their short‐maturity

counterparts.

Next, inspired by Zhang and Xiang (2008), we believe that the factors of the IV curve are good proxies of the risk‐

neutral moments. Therefore, they are expected to have predictability of the future excess returns of the underlying FXI

ETF, as is the case in other equity option markets (Ang, Hodrick, Xing, & Zhang, 2006; Chang, Christoffersen, & Jacobs,

2013; Chatrath, Miao, Ramchander, & Wang, 2016; Xing, Zhang, & Zhao, 2010). We test the predictability of FXI

monthly excess returns using the factors and the risk‐neutral third and fourth cumulants (Zhang, Chang & Zhao, 2019;

Ruan & Zhang, 2018), as well as their first differences (Ang et al., 2006) for the in‐sample and out‐of‐sample univariate

regressions. We find that the first differences of the third cumulants can predict the future FXI monthly excess returns

significantly in both in‐sample and out‐of‐sample tests.

Theories on the IV smirk have made vast progress in recent decades. Under Black and Scholes (1973), options with

the same time to maturity are supposed to have the same IV regardless of strike price. However, the IV calculated by the

standard Black and Scholes (1973) method was found to be different across strikes with the same underlying asset and

time to maturity (Rubinstein, 1985). Literature on the IV “smile”and “smirk”in the US market has been growing since

the initial study by Rubinstein (1985). Many studies have found that the phenomenon of the IV smile shape has become

a smirk shape since the global market crash in 1987; that is, the IV has become left‐skewed since then (Carr & Wu,

2003; Cont & Da Fonseca, 2002; Corrado & Su, 1997; Fajardo, 2017; Foresi & Wu, 2005; Skiadopoulos, Hodges, &

Clewlow, 2000; Yan, 2011).

To address the issue of different IVs at different strike prices, a number of stochastic volatility models have been

created (such as Barndorff‐Nielsen & Shephard, 2004; Bates, 1996; Heston, 1993; Stein & Stein, 1991). IV is useful to

measure the performance of a stochastic volatility option pricing model.

There is also a vast strand of literature trying to explain the causes of the shape of the IV curve (An, Ang, Bali, &

Cakici, 2014; DeMiguel, Plyakha, Uppal, & Vilkov, 2013; Garleanu, Pedersen, & Poteshman, 2009; Xing et al., 2010).

The errors of measurement and/or investor behavior are among the proposed explanations for the volatility skewness

(An et al., 2014; Bollen & Whaley, 2004; DeMiguel et al., 2013; Han, 2007; Hentschel, 2003; Pan, 2002; Xing

et al., 2010).

FIGURE 1 Performance of the FXI ETF and its benchmark index. It reflects the hypothetical growth of a $10,000 investment in the FXI

ETF and the benchmark index (ticker: XIN0I) from October 08, 2004 to April 30, 2018. Dividends are assumed to be reinvested. Fund

expenses are deducted for the FXI ETF. ETF: exchange‐traded fund

LI ET AL.

|

1451