Empirical study on the returns to contrarian investment strategies in Korean stock market

This study investigates whether contrarian investment strategies produce higher than normal returns in the Korean Stock Market, and, if they do, the reasons for such high returns. Contrarian investment strategies. call for buying stocks that have low prices relative to earnings and book assets (values?). While a consensus has been established that the strategies produce higher returns, the cause for higher returns are still controversial. This study analyzes descriptive evidences on the relation between the returns to contrarian strategies and variables associated with the naive expectations models (past sales growth rates and past earnings growth rates). It also provides formal tests of the naive expectations models and estimates the proportion of the returns to contrarian strategies by using Mishkin(1983)'s methodology. This study finds that the strategies produce 3-5% risk adjusted returns for the first year and 3-4% risk adjusted returns for the second year after the formation of the portfolio. Yet, this higher return cannot be explained by such risk elements such as beta and size. Therefore, this paper tests whether contrarian investment strategies are attributable for higher returns because the strategies exploit the suboptimal behavior of the typical investors. This study identifies three models of naive expectation formations that have been conjectured to explain the return to the contrarian investment strategies. Those three models are as follow: (1) prices reflect the naive extrapolation of past growth in corporate sales(2) prices reflect the naive extrapolation of past growth in corporate earnings, and (3) prices reflect analysts' forecasts of long-terms earnings growth. This paper finds that naive reliance on the past sales growth, rate can explain only about 20 similar to 25% of the higher returns to this strategy. However, earnings growth rate and analysts' forecast of future earnings growth rate do not offer substantial explanation for the higher return. This result is not consistent with Dechow and Sloan(1994) that found investors' naive reliance on analysts' forecasts of future earnings growth explains over half of the higher return to the contrarian investment strategies in the U.S. market. One reason for the difference in the cause of the higher returns between the two countries is that not like in the United States, the analysts' forecasts in Korea are not very actively produced and their credibility has not been firmly trusted among investors. Further, previous studies have shown that the analysts' forecasts are less accurate than the random walk model in forecasting earnings and sales before the Financial Crisis period in Korea. One limitation of this paper is that the behavioral hypotheses are developed on an ex-post basis to explain observed empirical regularities. The model could gain more credibility if the models are developed ex-ante on a theoretical basis.