ecent empirical research has documented many similarities between R stock market and futures market return seasonalities. Cross (19731, French (1980), Gibbons and Hess (1981), and Dyl and Maberly (1986) document a reliably negative Monday effect in both spot and futures stock index returns. Likewise, a negative growth rate in commodity futures prices on Monday is reported by Chiang and Tapley (1983) and Gay and Kim (1987). Rozeff and Kinney (1976) and Keim (1983) demonstrate that unusually high returns accrue to stocks during January. Gay and Kim (1987) also find that the monthly percentage changes in Commodity Research Bureau futures price index during January is higher than any other months' return. Although documentation of return seasonalities is extensive, researchers have been unable to identify their causes.' Ariel(1987) recently documented another curious anomaly in the monthly pattern of stock index returns: stock appears to earn positive returns only during the first half of calendar months, and zero average returns during the second half. Linn and Lockwood (1987) extended the investigation to the overthe-counter market and found that the same effect also prevailed in the daily NASDAQ composite index.
The objective of this note is to extend the investigation of the monthly effect to real goods markets. The hypothesis tested is that the average daily growth rates in raw material prices are the same between the first nine and the last nine trading days of the trading month.
Explanations for the observed weekend effect that have been examined and rejected include the delay between trading and settlement due to check clearing [Lakonishok and Levi (1982)1, measurement error in returns [Gibbons and Hess (1981) and Keim and Stambaugh (1984)1, and specialistrelated biases [Keim and Stambaugh (198411. Tax-loss selling pressure has been suggested as the cause of the January effect [Reinganum (1983) and Chang and Pinegar (198611, but the persistence of this phenomenon in some overseas markets with non-January tax year starting dates suggests that tax-loss selling only provides a partial explanation to the phenomenon [Brown, Keim, Kleidon, and Marsh (198311.