Price discovery among SSE 50 Index‐based spot, futures, and options markets

• Author: Yingyao Bi,Sungbin Sohn,Kwangwon Ahn
• DOI: http://doi.org/10.1002/fut.21970
• Date: 01 February 2019
• Published date: 01 February 2019

Received: 13 September 2018

|

Accepted: 15 September 2018

DOI: 10.1002/fut.21970

RESEARCH ARTICLE

Price discovery among SSE 50 Index‐based spot, futures,

and options markets

Kwangwon Ahn

1

|

Yingyao Bi

2

|

Sungbin Sohn

3

1

Graduate School of Future Strategy,

Korea Advanced Institute of Science and

Technology, Daejeon, Korea

2

Department of Derivatives Investment,

China Merchants Securities, Shenzhen,

China

3

HSBC Business School, Peking

University, Shenzhen, China

Correspondence

Sungbin Sohn, HSBC Business School,

Peking University, University Town,

Nanshan District, Shenzhen 518055,

China.

Email: sungbin.sohn@phbs.pku.edu.cn

Abstract

This paper studies the contribution of newly launched SSE 50 Index‐based

options and futures to price discovery. We find that the derivatives markets

quickly begin exhibiting price leadership over the corresponding spot market,

despite their short history; the information share from both derivatives markets

rose from 59.84% in mid‐2015 to 84.6% in mid‐2017. Using substantial

regulation changes during the sample period, we test the trading cost

hypothesis. The increases in derivatives transaction costs do not immediately

impede their roles in price discovery. Findings suggest that in nascent and

immature markets, investors’trading experience matters more than trading

costs.

KEYWORDS

Chinese derivatives markets, price discovery, trading cost hypothesis

JEL CLASSIFICATION

G13, G14

1

|

INTRODUCTION

The stock markets in mainland China, have achieved rapid development and recently became the second largest

following the US stock markets in terms of the market capitalization.

1

However, the country’s derivative markets are

still in their infancy. In April 2010, the China Financial Futures Exchange (CFFEX) launched the CSI 300 Index futures,

which is the first stock index futures in mainland China. In early 2015, the Shanghai Stock Exchange (SSE) and the

CFFEX launched the first exchange‐traded fund (ETF) options (SSE 50 ETF options) and the second index futures (SSE

50 Index futures), respectively, which marked the advent of a new derivative era in mainland China.

The introduction of futures and options goes a long way toward making up the limitations of trading mechanisms in

the current Chinese stock markets in several respects. First, derivatives virtually enable short‐selling, which is severely

limited in the underlying spot market. Second, derivatives are traded based on the “T+0”settlement system. In

contrast, regulators have used the “T+1”settlement system in the stock market since early 1995 to prevent excessive

speculation and potential flash crashes. Since the “T+1”system could aggravate speculation rather than prevent it

(Scheink & Xiong, 2003; Xiong, 2013), the “T+0”system in the derivatives markets can make information transmission

more efficient. Third, instead of the traditional continuous auction systems, the options market adopts the market‐

making system, which is expected to provide sufficient liquidity to the market. In theory, all of these aspects are

conducive to enhancing market quality (Gorton & Pennacchi, 1993; Subrahmanyam, 1991).

J Futures Markets. 2019;39:238–259.wileyonlinelibrary.com/journal/fut238

|

© 2018 Wiley Periodicals, Inc.

1

As of October 31, 2017, the market capitalization of the New York Stock Exchange and the NASDAQ is USD21,377 billions and USD9,585billions, respectively, while the Shanghai (Shenzhen) Stock

Exchange has the market capitalization of USD5,043 (3,688) billions.

A number of studies have documented the price discovery role between a single derivative market (either futures or

options) and its underlying spot market (an index or its tracker fund), showing that the derivative market responds

faster to the arrival of new information due to lower transaction costs and the lack of short‐sale restrictions

(Chakravarty, Gulen, & Mayhew, 2004; Chen & Chung, 2012; Chan, 1992; Chou & Chung, 2006; Iihara, Kato, &

Tokunaga, 1996; Kawaller, Koch, & Koch, 1987; So & Tse, 2004). However, relatively few papers have considered the

spot, futures and options markets jointly, particularly in emerging countries.

2

If both futures and options are actively

traded but one only considers a bivariate relationship between a single derivative market and its underlying spot

market, then the estimated contribution to price discovery can be biased because of the omitted variable. Since it is

unlikely that the contributions of the omitted market are exactly proportionally allocated to the other markets, the

bivariate analysis can be misleading. Not only does this paper fill the research gap between developed and developing

markets by using high frequency Chinese data, but it also estimates the relative contribution to price discovery in the

spot, futures, and options markets without potential bias by analyzing the three cointegrated markets jointly.

Compared to data used in previous studies, the sample in this paper is particularly interesting and valuable for two

reasons. First, the sample spans the period immediately after the launch of new futures and options products based on a

common index. This kind of serial launch is a rare event, which enables us to evaluate the progress of the derivatives

markets in mainland China. Unlike a mature and developed derivative market, a nascent and immature derivative

market particularly in an emerging country may not necessarily play its supposed price discovery role (Bae, Kwon, &

Park, 2009; Sohn & Zhang, 2017; Yang, Yang, & Zhou, 2012). From this sample, we can learn how soon nascent

derivatives markets function properly and what determines the strength of price discovery. Second and more important,

the Chinese derivatives markets have undergone dramatic up and down swings and abrupt changes in regulations

during this sample period. The transaction cost for futures trading was 0.0025% of the settlement amount at first, sliding

slightly to 0.0023%, rising five times to 0.0115%, skyrocketing 20 times to 0.23%, and then declining again to 0.092%.

Meanwhile, the option trading cost dropped by 35% from 0.02% to 0.013% of the transaction amount. The literature has

shown numerous evidence that supports the trading cost hypothesis (Booth et al., 1999; de Jong & Donders, 1998;

Fleming et al., 1996; Hsieh, Lee, & Yuan, 2008). In other words, the dominant price discovery role in the derivatives

markets is attributed to their relatively lower transaction cost. This sample period enables us to test this hypothesis

because substantial and almost exogenous changes in the transaction cost occurred within several years.

In this paper, we first provide an overview of the institutional facts about the launch of the SSE 50 ETF options and

the SSE 50 Index futures, after which we discuss their development over the 2‐year period. Using the Chow tests, two

structural breaks are detected in the sample period. The two identified structural breaks turn out to coincide well with

the major regulation changes. Next, we construct the comprehensive measures of the contribution to price discovery in

the spot, futures, and options markets, and investigate how they evolve over time. Specifically, we use SSE 50 ETF,

SSE 50 Index futures, and SSE 50 ETF options to represent the spot, futures and options markets, respectively, and then

quantify their price discovery strength by Gonzalo and Granger (1995)’s common factor weights (CFWs) and Hasbrouck

(1995)’s information shares (ISs), both of which are based on a cointegration relationship. We also consider the gross

and net spillover of each market by decomposing the forecast error variance attributed to innovations in each market

(Diebold & Yilmaz, 2012; Pesaran & Shin, 1998). Instead of a parametric specification of the time‐varying contributions

to price discovery, we estimate the vector error correction model (VECM) on a daily basis using intraday data to learn

its daily fluctuations as in Hasbrouck (2003).

We find that although the Chinese derivatives markets have only several years of trading history, they quickly began

exhibiting price leadership over the spot market. In the overall sample, the IS of the futures (options) market is 34.03%

(41.49%). The subsample analyses show that the leading price discovery role of the spot market quickly declines after

the launch of the two derivatives. In the first subsample, the IS of the spot market is 40.16%, but in the last subsample, it

is only 15.40%. The IS of the futures market has been large since the launch (37.54% in the first subsample; 30.17% in

the last), while the IS of the options market is mediocre in the first subsample (22.30%) but steadily increases and

becomes dominant in the last subsample (54.43%). The results are robust when the contribution to price discovery is

measured by the CFWs and the net spillover effects.

We also show that the bivariate analysis overestimate the contribution of the derivative markets to price discovery.

When we consider a bivariate relationship between spot and futures markets, the estimated IS in the futures market is

67.42%. For the same analysis with spot and options data, the options market takes 67.76% of the IS. In other words,

2

Booth, So, and Tse (1999) and Fleming, Ostdiek, and Whaley (1996) are two of a few papers that jointly examine the relative price discovery roles in the spot, futures and options markets. Their data

come from developed markets (the US or Germany) and the sample periods begin several years following the launches of the derivatives markets.

AHN ET AL.

|

239