اثر "ارزش" و بازار برای سهام چین

A long literature in empirical finance has isolated both a “value” and a small-capitalization effect in asset pricing. This study confirms the existence of these “style” effects both in new types of equity indexes and in the stocks of Chinese companies traded in international markets. We then present a new nonparametric method of portfolio construction that enables investors to extract the predictive power of these style effects, without diluting their efficacy through an unintended weighting distribution that closely resembles capitalization weighting. We then develop a simple method to isolate periods where style tilts are likely to be particularly effective.

A long literature in empirical finance has isolated a “value” effect in asset pricing. Studies such as Basu (1983) and Keim (1983) have shown that stocks selling at low prices relative to their earnings and book values have generated higher returns for investors. Similar results have been shown for stocks selling at low multiples to their sales. Fama and French (1992) confirmed a strong “value” effect in the United States stock market from the early 1960s through 1990. A particularly strong “value” effect characterized the U.S. stock market during the early 2000s as market prices adjusted from the levels that existed at the height of the “Internet Bubble.” Fama and French (1998) have also documented a strong “value” effect in international stock markets. One can interpret such findings as being inconsistent with efficient markets. Portfolios made up of stocks with low market values (MV) relative to book values (BV) earn excess risk-adjusted returns when risk is measured by beta from the Capital Asset Pricing Model (CAPM). But any test of market efficiency is a joint test of the relationship of returns to MV/BVs and the efficacy of CAPM's beta to fully measure risk. According to Fama and French, the ratio of market value to book value itself is a risk measure, and therefore the larger returns generated by low MV/BV stocks are simply a compensation for risk. Low MV/BV stocks are often those in some financial distress. Investigators such as Banz (1981) and Fama and French (1992) have also found a strong relationship between company size (measured by total market capitalization) and returns. Smaller firms appear to generate higher returns than larger firms. Again, the interpretation of these results is controversial. The excess returns of small firms can be interpreted as inefficiency, but they also may represent compensation for bearing risk. Smaller companies may be far more sensitive to economic shocks than are larger firms. Some studies of the stocks of Chinese companies over limited periods of time have confirmed the existence of style effects. For example, Wong et al. (2006) found that smaller firms and “value” stocks produced excess returns in the Shanghai Stock Exchange “A” share market over the period 1993 through 2002. Similar results have been reported by Bo and Krige (2008), Drew et al. (2003), Wong and DiIorio (2007), and Lam and Spyrou (2003). But as we have shown for the United States stock market, style effects are not dependably consistent.3Wong and DiIorio (2007) conclude that “there is no factor that has a persistent effect on stock returns.”4 There is also evidence that “momentum” strategies can yield excess returns in the Chinese market over the period 1995 through 2005.5Brown et al. (2008) find that both “value” and “momentum” strategies produced excess returns in four Asian markets (Hong Kong, Korea, Singapore, and Taiwan). They conclude, however, that a combination of the best value and momentum strategies does not provide a significant improvement over the best value strategy evaluated separately.

We have shown that “value” tilted portfolios appear to produce higher than market returns in the market for Chinese company stocks in most time periods from the late 1990s through mid-2008. But value-tilted portfolios do not consistently outperform capitalization-weighted portfolios. There appears to be evidence of mean reversion over time. Periods of lower relative returns for value-tilted portfolios often follow periods when value tilts have been effective. The rank method of portfolio construction described in this paper appears to be a particularly effective way to enhance the returns from a value style of investing. Rank weighting also appears to reduce the degree of mean reversion during periods when “value” stocks underperform the market. We have also shown that periods when value tilting is most effective correspond to periods when valuation metrics are very dispersed. The degree of compression of price-earnings multiples is a good predictor of the differences in returns between value-tilted and capitalization-weighted equity portfolios for a portfolio of the 25 largest H-share companies. The relationship is weaker, however, for an alternative set of 25 companies taken from the Hang Seng Index.