154 / 322
期貨未平倉量的資訊內涵及其交易活動之研究
154
The TAIFEX market structure differs from other developed markets in that transactions
are mostly contributed by retail traders. According to a statistical report on the TAIFEX
website, individual investors accounted for 85.49% of trading volume on January 2003,
whereas all institutional investors shared the remaining 15%. Retail traders are mostly short-
term speculators but rarely hedgers who hold futures positions for long. In addition, the
divergence/convergence in retail traders’ opinions would be different from that of
institutional traders. If open interest reflects the hedging demand as suggested by Chen et al.
(1995) or the divergence in traders’ opinions as in Bessembinder et al. (1996), the unique
market structure could have resulted in very different information contents of open interest in
Taiwan than in other markets. Kuo, Hsu, and Chiang (2005) study the Taiwan futures market
and show that the increase in expected open interest (viewed as market depth) does not
significantly mitigate volatility. Their results are not consistent with Chou and Wang (2006),
who find that open interest has a significantly negative impact on price volatility after the tax
reduction. This paper provides additional evidence on the informational contents of open
interest for Taiwan futures markets.
In next section, we construct three hypotheses regarding the information contents of
open interest, based upon existing literature. We provide data description and empirical
models in Section 3. In the fourth section, we present the empirical results for each
hypothesis. The last section concludes the paper.
2. Literature Background and Hypothesis Development
In this section, we develop three hypotheses and related implications to investigate the
information content of futures open interest. These hypotheses are generated through the
analysis of existing literature. We then develop testable implications under each hypothesis.
2.1 Market Participation
Hypothesis 1: Open interest reflects the participation of traders.
Based on the definition and the calculation of open interest, it represents the total
number of futures contracts that remains open and the amount of capital already flowed into
the futures market. Open interest is arguably an appropriate measure of current participant
activity, which is determined endogenously in the futures markets (Chang et al., 2000). If
open interest indeed reflects market participation, we would observe a close association
between open interest and liquidity measures, because more participation in trading usually