The Halloween Effect and Japanese Equity Prices : Myth or Exploitable Anomaly

Bouman and Jacobsen (2002) examine monthly stock returns for 37 world stock markets for the period January 1970-August 1998. They report that returns are significantly higher during the November-April periods versus the May-October periods in 36 of 37 markets examined and label this phenomenon the Halloween puzzle. These results conflict with those predicted by the efficient market paradigm. Using United States monthly return date, Maberly and Pierce (2004) report that, in terms of statistical significance, the Halloween puzzle is not robust to alternative model specifications. In particular, after adjusting for the October 1987 Crash and the August 1998 failure of Long-Term Capital Management, the Halloween puzzle becomes statistically insignificant at a meaningful level. Expanding on prior research, this paper examines the robustness of the Japanese Halloween puzzle to alternative model specifications and time periods. The data set is monthly Nikkei 225 index returns for the 1970 through 2003 period. The data set is divided into two sub-periods, 1970 through 1986 and 1987 through 2003, based on the internationalization of Japanese financial markets in the mid-1980s and the introduction of Nikkei 225 index futures in September 1986. For the entire period and for each sub-period, monthly returns are partitioned into bull market (positive return) years versus bear market (negative return) years. This classification reveals that the Halloween strategy underperforms the buy and hold strategy during bull market years, but outperforms the buy and hold strategy during bear market years. Thus, over any prolonged period of bull market years, the Halloween strategy fails miserably. The major finding of this study is that the Japanese Halloween puzzle is not robust to alternative model specifications or to the time period selected. Anecdotal evidence is presented suggesting that investment flows and trading volume in Japan display a November-April seasonal that might contribute to returns being numerically higher over the November-April periods. However, the numerically higher returns observed over the November-April periods are not economically exploitable nor are they a violation of the efficient market paradigm