Night trading and market quality: Evidence from Chinese and US precious metal futures markets

• Date: 01 October 2020
• Author: Neil Kellard,Ying Jiang,Xiaoquan Liu
• DOI: http://doi.org/10.1002/fut.22147
• Published date: 01 October 2020

J Futures Markets. 2020;40:1486–1507.1486

|

wileyonlinelibrary.com/journal/fut

Received: 19 March 2019

|

Accepted: 22 May 2020

DOI: 10.1002/fut.22147

RESEARCH ARTICLE

Night trading and market quality: Evidence from Chinese

and US precious metal futures markets

Ying Jiang

1

|Neil Kellard

2

|Xiaoquan Liu

1

1

Nottingham University Business School

China, UNNC-NFTZ Blockchain

Laboratory, University of Nottingham

Ningbo China, Ningbo, Zhejiang, China

2

Essex Business School and Essex Finance

Centre, University of Essex,

Colchester, UK

Correspondence

Neil Kellard, Essex Business School and

Essex Finance Centre, University of Essex,

Colchester CO4 3SQ, UK.

Email: nkellard@essex.ac.uk

Abstract

Given a dominant exchange, how should other exchanges set their trading hours?

We examine the introduction of a night session by the Shanghai Futures Exchange,

allowing trading concurrently with daytime trading at the Commodity Exchange in

the United States. After developing hypotheses, results for gold and silver show:

trading activity has increased; liquidity in Shanghai has risen and prices are less

volatile at market opening; the price discovery share of Chinese gold futures has

fallen but this is not a sign of weakening market quality; and volatility spillovers

increase bidirectionally. Longer trading hours have decreased market segmentation

and increased information flow.

KEYWORDS

information flow, intraday data, price discovery, trading hours, volatility spillovers

JEL CLASSIFICATION

G13; G14

1|INTRODUCTION

The concept of dominant and satellite markets was first proposed by Garbade and Silber (1979) who analyze the short‐run price

behavior of an identical asset traded in two different markets. Since then, numerous studies have investigated price discovery in

informationally linked markets (see H. Chen, Choi, & Hong, 2013;Gonzalo&Granger,1995;Hasbrouck,1995; Rapach,

Strauss, & Zhou, 2013;Ye,2014). In global futures trading, it has been found that preeminent exchanges take the lead in price

discovery for certain commodities (see Covrig, Ding, & Low, 2004;Gong&Zheng,2016; Han, Liang, & Tang, 2013;Jiang,Su,

Todorova, & Roca, 2016;Liu&An,2011;Yin&Han,2013). Given this, how should newer entrants behave, particularly when

setting trading hours? Clearly, there may be liquidity shortages if a market is open for long periods during which demand for

trading is scarce. On the other hand, a shorter trading period mayattractenoughvolumebutgivesrisetomoretimewhenthe

market is closed. New information impounded in the price of the preeminent exchange might not be reflected in the price of

the entrant for several hours, leading to price jumps and unnecessary volatility when the entrant market opens.

To investigate the question of optimal trading hours for newer entrant markets, we examine the introduction of

night trading by the Shanghai Futures Exchange (SHFE) for gold and silver futures

1

on July 5, 2013. Before this period,

-------------------------------------------------------------------------------------------------------------------------------------------

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided

the original work is properly cited.

© 2020 The Authors. The Journal of Futures Markets published by Wiley Periodicals LLC

1

As additional motivation note that the price behavior of international commodity futures markets, including precious metals, has been shown to be

useful in predicting business cycles (see Kang & Kwon, 2017) and stock market returns (see Alvesy & Szymanowska, 2019) in many countries.

trading took place during daylight hours in Shanghai between 9:00 a.m.–11:30a.m. and 1:30 p.m.–3:00 p.m. From July 2013, an

additional session was added which runs from 9:00 p.m. to 2:30 a.m. the following day (see Jin, Li, Wang, & Yang, 2018), more

than doubling the amount of time available for trading during a ‘trading’day. The SHFE view was, “The purpose of night

trading is to help prices [on the Shanghai exchange] better connect with global prices, and to help achieve our goal of

internationalizing our contracts”(Li, 2013). So did this actually happen? And, in particular, what happened to important

measures of market quality, such as volume, liquidity, price discovery, and volatility, as a result of this new policy?

Although the SHFE is the second largest gold futures exchange worldwide by volume traded (Jin et al., 2018;

Reuters, 2017), the Commodity Exchange (COMEX) in New York, where gold futures first started trading in 1974, is

larger. Indeed, recent work suggests that the COMEX gold futures play the leading role globally in price setting and

discovery (Fung, Tse, Yau, & Zhao, 2013; Hauptfleisch, Putnins, & Lucey, 2016; X. Xu & Fung, 2005). By opening at

9:00 p.m. in China, which is 9:00 a.m. Eastern Standard Time (EST) in New York, the SHFE trading for gold and silver

futures now overlaps with the active trading period in the United States given that the COMEX opens at 8:20 a.m. EST.

To examine the effect of this overlap, our paper develops several testable hypotheses which we subsequently

empirically test using both daily and intraday data from Chinese and US futures markets. Specifically, we posit that post

the introduction of night trading at the SHFE, in Chinese markets (a) volume will be higher at night; (b) liquidity will

rise; (c) perhaps counterintuitively, price discovery measures will fall; and (d) volatility spillovers will increase bidir-

ectionally but particularly from the United States to China.

Employing a sample period from January 2008 to April 2016, our empirical tests validate our hypotheses to a large

extent but also reveal some unexpected findings. First, trading activity, as measured by volume, turnover, and open

interest, for both China and the United States is higher after the introduction of night trading. Moreover, after the

introduction and in China at least, such activity is higher during the night trading session than the daylight alternative.

This provides a prima facie case that Chinese futures traders value the night trading session and that, consequently,

market quality indicators are likely to have improved.

Second, using three popular measures, namely, Roll's spread, the Amihud illiquidity measure, and the proportion of

zero returns, we examine liquidity dynamics before and after the introduction of night trading. The Roll and Amihud

measures, specifically, indicate that typically liquidity has improved at the SHFE as hypothesized postintroduction and

suggest that market quality has increased. On average, traders can make transactions more easily and with less price

effects than before July 2013. In particular, comparing the first daylight trading hour in China pre‐and postnight

trading, Roll, Amihud and realized volatility measures are lower for the postnight trading subsample. The opening of

the Chinese night trading session, crucially overlapping with the COMEX daylight trading period, enables the SHFE

markets to capture relevant price information from the United States in real time. As a consequence, this ensures a less

volatile market when Shanghai opens the following morning.

Third, recent literature suggests that price discovery is more likely to occur in the SHFE gold futures market relative

to the Shanghai Gold Exchange (SGE)

2

physical market and that, within both markets, price discovery is more

prevalent during night trading sessions (Jin et al., 2018). Here we extend such work by examining the price discovery

function of the SHFE gold and silver futures relative to the preeminent US futures exchange. Strikingly, employing

established measures, such as the component share (CS) of Gonzalo and Granger (1995), the information share (IS) of

Hasbrouck (1995), and the modified information share (MIS) of Lien and Shrestha (2009), we show that for gold, price

discovery share actually falls relative to the United States after the introduction of night trading. However, we posit this

is not a sign of weakening market quality but the contrary. Before July 2013, Chinese and US markets were more

segmented, as the SHFE was closed during the active daylight trading period in the United States. However, night

trading in China now allows greater information transmission between the two markets over a much longer, con-

temporaneous trading period. This lessening of market segmentation implies that the preeminent market's share of

price discovery will rise and this is observed empirically.

3

Finally, to further investigate the changing interdependence between the two exchanges we implement the volatility

spillover index proposed by Diebold and Yilmaz (2012). As hypothesized, directional (and total) spillovers grow fol-

lowing the introduction of night trading but particularly from the United States to China for both commodities.

Moreover, while volatility spillovers increase most starkly from the United States to China during night trading, they

2

Other work examining market quality at the SGE includes C. Xu and Zhang (2019).

3

Interestingly, price discovery measures for Chinese silver futures rise after the introduction of night trading, albeit from a very low base. This we

ascribe to the SHFE silver futures market being particularly new, opening as it did in June 2012, only a year before night trading was introduced. US

silver futures still dominate in terms of price discovery in postnight trading.

JIANG ET AL.

|

1487