A rare move: The effects of switching from a closing call auction to a continuous trading

• DOI: http://doi.org/10.1002/fut.22081
• Author: Ya‐Kai Chang,Robin K. Chou,J. Jimmy Yang
• Published date: 01 March 2020
• Date: 01 March 2020

J Futures Markets. 2020;40:308–328.wileyonlinelibrary.com/journal/fut308

|

© 2019 Wiley Periodicals, Inc.

Received: 3 January 2019

|

Accepted: 21 November 2019

DOI: 10.1002/fut.22081

RESEARCH ARTICLE

A rare move: The effects of switching from a closing call

auction to a continuous trading

Ya‐Kai Chang

1

|

Robin K. Chou

2,3

|

J. Jimmy Yang

4

1

Department of Finance, College of

Business, Chung Yuan Christian

University, Taoyuan City, Taiwan

2

Department of Finance, National

Chengchi University, Taipei City, Taiwan

3

Risk and Insurance Research Center,

National Chengchi University,

Taipei City, Taiwan

4

College of Business, Oregon State

University, Corvallis, Oregon

Correspondence

Ya‐Kai Chang, College of Business, Chung

Yuan Christian University, 200 Chung Pei

Road, Jhongli, Taoyuan, Taiwan.

Email: ykchang@cycu.edu.tw

Funding information

Ministry of Science and Technology of

Taiwan, Grant/Award Number:

105‐2410‐H‐033‐002; National Natural

Science Foundation of China,

Grant/Award Numbers: 71772156, 71973018

Abstract

This study investigates the effects of switching to a closing continuous trading

(CCT) on market quality, while considering the trading behaviors of different

types of traders. Investors become more patient in the period preceding the last

trading phase, whichreduces the bid–ask spread (BAS) inthat period. We find an

increase in the BAS and volatility during the last trading phase, due to

diminishing investor patience. Market volatility and the closing pricing errors

relate positively to the trading activities of foreign institutional investors. Overall,

the introduction of the CCT worsens the market quality before the closing.

KEYWORDS

closing call, continuous trading, market quality, price discovery, trading activity

JEL CLASSIFICATION

G14, G15, G18

1

|

INTRODUCTION

Exchanges constantly seek improved closing trading mechanisms to render asset closing prices that better reflect

fundamental values. However, disagreements and variations arise regarding the use of either a closing call auction

(CCA) or continuous trading at the close; determining the preferred closing method is an important task. Existing

theoretical literature has yet to provide a clear answer on this issue. Empirical studies also are lacking, due to the

infrequency of changes to the closing mechanisms. In particular, theoretical and empirical literature regarding the

effect on market quality of switching from a CCA to a closing continuous trading (CCT) mechanism remains scarce. A

rare move—as exhibited by a switch from a closing call to a CCT mechanism on the Taiwan Futures Exchange

(TAIFEX)—enables us to provide empirical evidence about the changes in market quality before and after the switch,

which sheds light on the relative performance of closing mechanisms. Furthermore, by considering the effects of the

trading behavior of different types of traders, our findings have important policy implications for regulators who are

especially concerned about the welfare gains and losses of different types of market participants due to the closing

mechanism changes.

Finding a closing trading mechanism with efficient closing prices is even more important for futures markets, due to

their mark‐to‐market system. Existing evidence of the effectiveness of either call auction or continuous trading as a

closing trading mechanism focuses primarily on stock markets rather than on futures markets though. Most

importantly, this article presents an interesting test to examine whether and why market participants might have

“precursor”reactions to the closing mechanism changes, as well as whether and how different types of traders adjust

their trading to the new closing environment. In so doing, this study attempts to fill a gap in existing literature with a

more complete analysis of whether, why, and how a rare switch from a CCA to the CCT might affect the trading

behavior of different types of traders and thereby influence market quality.

In a call market, buy‐and‐sell orders accumulate over a specified time interval for simultaneous execution in a

batched trade, at a single price. In a continuous market, a trade instead may occur at any time that a bid and offer

match or cross. Multiple theoretical models (e.g., Budish, Cramton, & Shim, 2014, 2015; Garbade & Silber, 1979;

Glosten & Milgrom, 1985; Ho, Schwartz, & Whitcomb, 1985; Madhavan, 1992; Mendelson, 1982, 1987) concur that call

markets can mitigate information asymmetry problems, but they lack the advantages of continuous markets, including

immediacy in trading and lower costs for gathering market information.

Empirically, the effects of changes in closing mechanisms appear mixed. Some studies find that the introduction of a

closing call procedure improves the price discovery process and market quality (Chang, Rhee, Stone, & Tang, 2008;

Chelley‐Steeley, 2008; Comerton‐Forde, Lau, & McInish, 2007; Ibikunle, 2015; Kandel, Rindi, & Bosetti, 2012; Pagano &

Schwartz, 2003).

1

However, Sivanithy and Ng (2000), Comerton‐Forde and Rydge (2006), and Chelley‐Steeley (2009) find

no such evidence.

2

Similarly, mixed results emerge from the introduction of a CCT. Whereas Amihud, Mendelson, and

Lauterbach (1997), Lauterbach and Ungar (1997), Muscarella and Piwowar (2001), and Kalay, Wei, and Wohl (2002)

provide evidence that a CCT is superior to a CCA trading, Lauterbach (2001) suggests that its introduction may not

improve market quality for illiquid stocks.

3

The effects of the changes of closing mechanism thus are far from conclusive.

An efficient closing mechanism is important to both market participants and regulators, because it confirms that the

closing price will reflect the fundamental values of the asset. Many market participants, especially mutual funds, have

strong incentives to trade at closing prices. The settlement with the departing (or new) investors is based on the closing

price, so mutual funds that handle investors’redemption and injection of funds prefer to execute their trades as close to

the closing price as possible, to avoid settlement risk. This potential demand for trading at the closing price results in

liquidity shocks at the end of the trading day and induces large futures price movements, which cause closing prices to

move away from fundamental values (Cushing & Madhavan, 2000; Hagströmer & Nordén, 2014). A closing mechanism

that can promote price discovery thus would be desirable.

An efficient price at the market close is critical because closing prices are widely cited. Comerton‐Forde and Rydge

(2006), Kandel et al. (2012), and Hagströmer and Nordén (2014) indicate that closing prices are used to evaluate

portfolio returns and measure mutual funds’net asset value. Moreover, closing prices serve to determine the value of

other derivative contracts and benchmark the performance of portfolio managers. The closing price on futures markets

also provides a basis for measuring whether investors’margin requirements are met on futures markets. The mark‐to‐

market system computes investors’daily profits or losses, based on the daily closing price. If the current market value

causes the margin account to fall below its required level, the investor will face a margin call.

Although the closing trading mechanisms are important for futures markets, most studies focus on stock markets. Existing

literature usually examines a switch from the CCT to the CCA, but empirical evidence of the shift from a call auction to a CCT

is scarce, due to the rarity of such events. To our knowledge, the Tel Aviv Stock Exchange (Amihud et al., 1997; Kalay et al.,

2002; Lauterbach, 2001; Lauterbach & Ungar, 1997) and the Paris Bourse (Muscarella & Piwowar, 2001) are the only two

markets that have experienced such a move. No prior studies cover futures markets that move from a CCA to a CCT.

But on October 8, 2007, the TAIFEX switched from a CCA to a CCT for the last 5 min of trading, creating a good

opportunity to study the effects of the closing mechanisms of futures markets. The TAIFEX is an order‐driven electronic

futures market with no designated market makers. Trading on the TAIFEX occurs between 8:45 and 13:45.

4

According

to the TAIFEX, the CCT mechanism was attached to the introduction of a futures calendar spread mechanism,

1

Consistent with these findings, Schwartz (2001) notes that the CCA enhances price efficiency, lowers price volatility, and narrows bid–ask spreads (BASs). Kandel et al. (2012) find a significant

reduction in spread and volatility and an increase in the aggressiveness of liquidity suppliers and price efficiency 10 min before the CCA. Chelley‐Steeley (2008) and Ibikunle (2015) also show that the

least active securities in the London Stock Exchange experience the greatest improvements to market quality following the introduction of a closing call.

2

Sivanithy and Ng (2000) and Comerton‐Forde and Rydge (2006) demonstrate that the implementation of a CCA does not completelyremove market manipulation. Comerton‐Forde and Rydge (2006)

also indicate that matching criteria used to set auction prices can greatly influence the effectiveness of the closing calls. Finally, Chelley‐Steeley (2009) shows that a CCA may not enhance the market

quality of the London Stock Exchange for high liquidity securities.

3

Amihud et al. (1997), Lauterbach and Ungar (1997), and Kalay et al. (2002) indicate that CCT is superior to a CCA in the Tel Aviv Stock Exchange. In addition, Muscarella and Piwowar (2001) find

that market quality increases at the Paris Bourse for frequently traded stocks that are transferred from their call market to their continuous market. Despite the aforementioned benefits, Lauterbach

(2001) offers a different insight and shows that not all stocks benefit from a continuous trade. For small and infrequently traded stocks, a switch from a call auction to continuous trading may not be

optimal.

4

Given that the TAIFEX closes 15 minafter the spot market, our results could potentially be biased and thus should be interpreted with caution. Nevertheless, assuming that any potential biases

caused by futures market late closures are similar for boththe CCT‐regime and the CCA‐regime,our results of comparing the market quality before and after the new closingmechanism would remain

valid, because we focus on the relative rather than absolute changes.

CHANG ET AL.

|

309