Analysis of the overreaction effect in the Chinese stock market

Abstract

Several recent studies have examined whether the main Chinese stock markets in Shanghai and Shenzhen are weak-form efficient. A consistent feature of the findings is that the pricing of foreign-owned B shares is more predictable than domestically-owned A shares. However, none of the earlier investigations examine the overreaction effect, one of the most commonly-employed tests of weak-form efficiency in developed stock markets. The present study therefore reports the results of such an analysis for a sample of more than 300 Chinese shares over a six-year period beginning in August 1994. In contrast to earlier evidence, the article finds that the overreaction effect is most pronounced in the market for A shares, suggesting that the normal impression of greater efficiency in the pricing of Chinese-owned equities may be open to further challenge and debate.

Notes

¹ The weak-form of the efficient markets hypothesis (EMH) can be traced back to Fama (1965). The EMH states that a market is efficient (in the weak-form) if it cannot be outperformed consistently by trading on the basis of historic information. Although most early studies concentrated on the US and UK markets (Fama and Blume, 1966; Cootner, 1967; Niarchos, 1972), recent years have seen the focus shift to markets in developing countries (Bessembinder and Chan, 1995; Coutts and Cheung, 2000). For a review of the literature on emerging markets see Gunasekarage and Power (2001).

² The Shenzhen and Shanghai Stock Exchanges opened on the 1 and 19 of December 1990 respectively. By the end of 2000, the combined market capitalisation of the exchanges was US$581bn, split almost equally between the two. In the first ten years following the opening of the markets, China's market capitalisation to gross domestic product (GDP) ratio grew from zero to 51%.

³ That is a trading rule is established on the basis of buying a share when it rises 45% above a previous low, holding it until the price falls 45% from a previous high and then adopting a short position until a further buy signal is generated. See Fama and Blume (1966) for further details about this technique.

⁴ The alteration to the rules took place on the 28 February 2001; see McGuinness (2002). In December 2002 the rules concerning the ownership of A shares were relaxed to allow limited participation by foreigners. The A share market was opened to international institutional investors under certain conditions, including a requirement for each Foreign Qualified Institutional Investor to have funds under management of at least $10bn and to keep any capital invested and any related income within China for 3 years.

⁵ The main Chinese markets are regulated by The China Securities Regulatory Commission (CSRC). Following a series of rationalisations of supervisory activity, the CSRC became the sole body responsible for market regulation in April 1998 and reports directly to the State Council.

⁶ As Wang et al. (2003) argued, the move to widen ownership of B shares may have been predicted by market participants prior to the formal rule amendment taking effect at the end of February 2001. The present study therefore follows Wang et al. (2003) and employs a data set ending several months prior to the change.

⁷ A total of 15 business sectors are represented in the sample, although the ‘Heavy Industry’ sector dominates, with 127 firms.

⁸ See Strong (1992) for a discussion of the strengths of the market-adjusted model. Its use in this case reflects the likely existence of thin-trading and beta instability in many Chinese equities (Bailey, 1994, Ma, 1996) which can affect the robustness of results derived from more sophisticated return-generating models.

⁹ Poon et al. (1992) and others suggest that around 25 shares are needed in a portfolio to diversify away unique risk.

¹⁰ Although not reported here, the six-year period was divided into 12 six-monthly sub-periods and the overreaction tests reperformed. The results indicated that the strength of the overreaction effect declined markedly in the latter half of the period. This finding was not unexpected, given on-going regulatory changes in the Chinese capital market, the inflow of foreign investment and greater investor scrutiny in recent years (Fernald and Rogers, 1998; Su and Fleishcer, 1998). These results are available from the authors on request.