Academic journal article Journal of Economics and Finance

Value Stocks and Market Efficiency

Academic journal article Journal of Economics and Finance

Value Stocks and Market Efficiency

Article excerpt


We form portfolios based on firm book-to-market equity ratios and apply stochastic dominance tests. Value (high bookto-market) portfolios dominate low book-to-market portfolios. Thus, value stocks are not rationally priced by the market and the book-to-market ratio is not an efficiently priced proxy for equity risk. We also find that the superior performance of value stocks is not due to the January effect. (JEL G140)


The superior return performance of value stocks as characterized by high book-to-market equity (BEME) ratios is now well documented. Fama and French (1992) find that the return on high BEME stocks is, on average, greater than that of low BEME stocks. This result holds true across different time periods (Davis 1994), for foreign stocks (Chan, Hamao, and Lakonishok 1991), and for a holdout sample of financial firms (Barber and Lyon 1997).

Controversy exists on whether the superior returns of value stocks represent an exception to the efficient market hypothesis (EMH). Fama and French (1992, 1993) suggest that the BEME ratio is a proxy for some presently unknown risk factor. In this case, the large returns observed in high BEME stocks would be compensation for bearing risk and, thus, would be consistent with market efficiency. Other researchers, however, contend that the performance of value stocks violates rational pricing. Haugen (1993); La Porta, Lakonishok, Shleifer, and Vishny (1997); Lakonishok, Shleifer, and Vishny (1994); and Shefrin and Statman (1995) all argue that investor expectations concerning value stocks are systematically biased.

We attempt to resolve the controversy using stochastic dominance (SD) analysis. We form stock portfolios based on ranked values of the BEME ratio and apply SD tests to determine if preferences (dominance) can be established among the portfolios. If value stocks are rationally priced and the market is efficient, no preferences should be established among high-BEME portfolios and low-BEME portfolios. This is true even if the equilibrium, return-generating process is complex and poorly understood. The failure of SD analysis to establish preferences would imply that the BEME ratio is a proxy for some unidentified equity risk factor. If, however, preferences can be established on the basis of the BEME ratio, then value stocks are not rationally priced, and the market is not efficient.

Stochastic dominance has important advantages over traditional market efficiency tests. Traditional tests of market efficiency compare observed returns to expected returns generated by some equilibrium model, such as the capital asset pricing model or the market model. Thus, abnormal returns indicative of market inefficiency may, in fact, be due to omitted risk factors or other pricing model misspecification. In contrast, stochastic dominance tests examine the entire distribution of returns, eliminating the need to identify and mathematically formalize the return generating process. Falk and Levy (1989) discuss the rationale underlying stochastic dominance tests of market efficiency in more detail.

Our basic findings are straightforward. High-BEME portfolios dominate low-BEME portfolios. The performance of value stocks is a puzzling challenge to market efficiency. We also find that the superior performance of value stocks is not due to the January effect.

Stochastic Dominance

Stochastic dominance provides an effective method for placing investment choices into mutually exclusive sets: an efficient set containing desirable investment alternatives and an inefficient set containing undesirable ones. Return distributions of portfolios formed on the basis of the BEME ratio are compared to one another to determine which would be preferred (i.e., in the efficient set versus not in the efficient set). The preference criteria are that investors prefer more return per dollar invested to less, are risk averse, and prefer positive skewness (potential for great gain with limited downside risk) in a return distribution. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.