زمانی که بازار نقد زمینه ای، معامله نمی کند، چه کسی در بازار اتی شاخص سهام معامله می کند؟

This paper studies the trading of Hang Seng Index futures contracts on the Hong Kong Futures Exchange since the Exchange began to open earlier and close later than the underlying cash market by 15 min. We show that the extension of trading hours stimulates the opening trading volume of the futures market. Futures returns surrounding the market opening are found to be relatively less volatile with insignificant change in pricing errors when compared with the pre-extension period. These observations suggest that activities during the extended opening session are related with private information but not with public information or noise.

The study of interdependence between futures and cash markets has attracted much attention from finance academics and practitioners.1 Although most of these studies have focused on institutional settings in which both futures and their underlying cash markets are concurrently traded, the behavior of a futures market under the closure of its underlying cash market is another interesting topic in the finance literature. Among the extant studies in the latter line of research, Chang et al. (1995) conducted a pioneer empirical investigation on Standard and Poor's (S&P) 500 index futures market during the last 15-min period when the underlying stock market is closed. Their findings confirmed the prediction of certain financial models. Specifically, they found that the volatility of futures prices first drops, as predicted by the contagion model of King and Wadhwani (1990), and subsequently increases in the closing minutes, as suggested by the asymmetric information models of Admati and Pfleiderer (1988) and Foster and Viswanathan (1990). Following Chang et al. (1995), several studies were conducted on the Hang Seng Index (HSI) futures contracts traded on the Hong Kong Futures Exchange (HKFE). Ho and Lee (1998) showed the presence of contagion and information effects for the period from April 1993 to March 1997 when the HSI futures market extended its trading hours beyond that of the underlying cash market by 5 to 15 min. Fong and Frino (2001) suggested that the decline in the HSI futures volatility associated with the closure of the cash market can partially be explained by the fall in the bid–ask bias of futures prices. In a more recent study, Cheng et al. (2004) showed that return innovations from the extended opening session of the HSI futures market help to explain the overnight return in the cash market. Despite the existence of the previously mentioned studies, an important research issue is still waiting to be addressed: which type of investors dominates the trading of futures contracts when there is no activity in the cash market? Volatilities in securities market can generally be classified as those originated from public information, private information, or noise (French and Roll, 1986). Although information-based trading helps to speed up the price discovery process, noise trading causes the price to deviate from its intrinsic value. By investigating which type of investors is trading, we can understand whether the price contained in futures contracts is informative or noisy when the cash market is closed. This paper makes a contribution to the existing literature by formulating and testing the public information, private information, and noise trading hypotheses for the extended trading hour of Hong Kong's HSI futures market. At the same time, it also differs from the pervious related studies in several other aspects. Chang et al. (1995) and Ho and Lee (1998) examined only the last 5 to 15 min of the futures market after the cash market has closed. Our study provides a richer investigation as no concurrent trading occurs of the underlying stocks for both the first and the last 15 min of the futures market during our sample period. The extended opening session of the futures market is characterized by the accumulation of information arrived overnight, and we are interested whether it will behave differently from the extended closing session. Furthermore, we analyze the pricing efficiency of future contracts—an issue that has not been addressed in any previous study. On November 20, 1998, the HKFE extended the trading hours of its HSI futures contracts by opening 15 min earlier and closing 15 min later than the underlying cash market, the Stock Exchange of Hong Kong (SEHK).2 The major impact of this event can be briefly explained as follows. The extension of trading hours stimulated the opening trading volume of the futures market but the opening futures returns are found to be relatively less volatile than before. In addition, no significant changes occurred in pricing errors when compared with the pre-extension period. These observations reject the hypotheses that activities during the extended opening session of the futures market are primarily related to public information or noise, but they cannot reject the private information hypothesis. On the other hand, no notable changes exist in return volatility, trading volume, and pricing efficiency in the last 15-min trading session of the HKFE during the post-extension period. Thus, evidence is found for the time-of-the-day cash market closure effect on futures trading. The remainder of the paper is arranged as follows. The next section describes the data used in this study. Section 3 develops the empirical implications of the public information, private information and noise trading hypotheses. Section 4 gives a brief description of the measurement of volatility, volume, and pricing efficiency. The empirical results are presented in Section 5. The last Section concludes the paper.

This paper examines the effects of the extension of trading hours in the Hong Kong futures market on the behavior of HSI futures contracts. The empirical results indicate increases in trading volume, reductions in the volatility of futures returns, and insignificant changes in pricing errors occur during the extended 15-min opening session. These findings are consistent with the hypothesis that increased private information-based tradings in the futures market occur before the opening of the underlying cash market. However, no significant changes in trading behavior are found for the last 15-min closing session during the post-extension period. Although this study shows that investors with private information trade heavier on the futures market before the cash market has opened, it will be worthwhile to investigate the rationale for adopting such a trading strategy. One possible approach is to study the equilibrium trading strategy for this type of trader when both the cash and futures markets are open and analyze how the equilibrium changes when one of the markets is closed. Such a study could prove an interesting topic in research on market microstructure.