opening call auction - pre-market

The following is a precis of a research paper by Department of Finance Discipline, University of Sydney, NSW, 2006, Australia. For readability, it has been heavily sub-edited to reduce it to its essentials. Tables, supporting references, and arithmetic formula have been removed. Some academic language has been replaced with every-day language.

The original un-abridged paper, commissioned by the SFE can be seen at  opening auction - University of Sydney

The "privileges" granted by exchanges to market-makers are covered in earlier works. It is unclear from this paper whether the "monopoly power" is a given privilege and an acccepted fact, or the assertion is a product of their research. We are more interested in the mechanics of how the market-maker achieves its "monopoly power".

footnote
In 2001 the SFE trialled an extended day-session, pre-open at 09:20am, trading commenced 09:30am, with a 30 minute gap until the physical opened at 10:00am. Trading during the extended 20 minutes was volatile. The trial was discontinued after 6 months. No prizes for guessing why.

legend
specialist = market-maker
open to open = open price previous day to opening session price today
close to close = EOD closing price Previous Day to EOD closing price current Day
Inter-Day = period spanning several days
Intra-Day = within the current day


This research paper by University of Sydney, was commissioned by the Sydney Futures Exchange. Refer opening credits in the unabridged paper above.


source
ASPRIS, FRINO, LEPONE
Department of Finance Discipline
Faculty of Economics and Business
University of Sydney, NSW, 2006, Australia.

This paper examines the opening call auction mechanism in futures markets. We examine a controversial issue surrounding the reasoning of why open to open returns exhibit greater dispersion than close to close returns and present evidence regarding the opening price mechanism of nine international derivative exchanges. We support the proposition that the long halt in trading before the market opening results in greater uncertainty and more volatile prices. We reject earlier findings that suggest that this result is driven by the liquidity of the contracts surveyed.

Introduction
Despite extensive research examining inter-daily return volatilities, there is little consensus amongst academics and market designers why open to open returns exhibit greater dispersion than close to close returns. Amongst the reasons advocated are alternative trading mechanisms between the market’s open and close, the long halt in trading before the market opening.

This study contributes to previous studies examining this issue across nine stock index futures contracts with varying opening market structures. Previous studies have specifically focused on US equity markets and results have consistently indicated the presence of greater volatility at the open. This study differs from existing studies in that we examine the issue within a futures market framework, where contracts are inherently more liquid than stocks traded in equity markets.

By examining only international index futures contracts we are able to test that the presence of a higher open-to-open volatility component is related to the liquidity of the stock or contract. Furthermore, the sample of exchanges tested includes contracts that trade in one or multiple trading session, which makes it possible to test the proposition that the higher volatility at the open of the trading day is a function of the preceding non-trading period rather than the trading mechanism.

Literature Review
Early studies conducted on NYSE and other international stock exchanges indicate that open to open return volatilities are up to 20% higher than close to close return volatilities. This aberration in return volatilities is ascribed to the different trading mechanisms used at the open and close. However, it has also been suggested that open to open returns display greater volatility due to the ability of the specialist (market maker) to exploit their monopoly power at the open. The market-maker is able to observe the order book prior to setting the opening price which allows him to earn a monopoly profit that is not possible during the rest of the day.

Research that has examined this indicates that this result is not restricted to markets that open with a call-auction. Using the Italian market as an example, the open return volatility is higher when the market opens in a continuous rather than a call auction manner. This introduces a third intervention which separates further the trading mechanism from the non trading hours.

In order to recognize the differing effects, examination of the Japanese market which has two daily trading sessions, both characterised by a call auction at the open and a continuous session to the close, disclosed that the morning open to open return volatility is higher than the close but not the afternoon open to open volatility supporting that the anomaly is a function of the preceding non-trading period rather than the trading mechanism. After introducing a 30 minute pre-trading routine prior to the market’s opening, opening volatility fell by over 60%. This suggests that minimising the period of non-trading or lengthening the period of price discovery in the pre-opening period may mitigate issues relating to the inconsistency between open to open and close to close return volatilities.

Additional results in this matter are however, mixed. Examination of the Stock Exchange of Hong Kong (SEHK) which is an order-driven market with no specialist system shows that open to open volatility is slightly lower than the close to close volatility. Their results are similar to those who suggest that monopoly pricing is the reason for the difference in volatilities. Examination of the Japanese market using the TOPIX index return similarly does not observe higher open to open return volatility. On the other hand a study of the Korean market found that both the morning and afternoon opening volatilites are significantly higher than the market closing volatility even though auction clearing is used in all these occasions.

Data and Methodology
The data in this study is sourced from Reuters and contains all transactions in 9 stock index futures contracts from January 1, 2005 to December 31, 2006. The sample is restricted to electronic trading in the near contract during daytime trading hours. The importance of the opening is depicted by the trade volume executed at the open which we present in Table 1. We find that on average 3% of daily volume is executed using the opening price mechanism.

Examination of the ratios, indicates that traders who choose to transact at the opening are exposed to a higher risk and its associated costs. As highlighted earlier, there is a mixture of findings surrounding the inter-daily volatility component.

This analysis is contradicted by analysis, which shows that while a positive volatility in the morning opening period exists on the Japanese market, the afternoon opening period does not exhibit similar traits. Our results are consistent with the proposition that longer halts between sessions result in greater uncertainty and volatility in prices. We find that in the three markets with afternoon trading sessions that the opening variance ratios associated with the afternoon trading period are either less or not significantly different. This combined with the fact that most exchanges exhibit a greater transitory volatility component in the morning opening session is consistent with the long halt in the trading process.

We find no evidence that the presence of a higher open-to-open volatility component is related to the liquidity of the stock or contract. We show that these inherently liquid contracts exhibit a higher opening volatility in the morning sessions.

Conclusion
This paper presents evidence regarding the opening price mechanism of international derivative exchanges, with specific references to futures index contracts. We find the long halt in trading before the market opening results in greater uncertainty and more volatile prices. We reject any findings that suggest that this result is driven by the liquidity of the contract through the inherent dimensions of the contracts surveyed.


camron.systems.australia 2007