Resumen de Comparison of the �turn-of-the-month� and lunar new year return effects in three Chinese markets: Hong Kong, Shanghai and Shenzhen
Paul B. McGuinness, Richard D. F. Harris
Within the context of the mainland Chinese (Shanghai and Shenzhen) and Hong Kong market places, we investigate two of the most important documented calendar anomalies: the �turn-of-the-month� and Chinese Lunar New Year (CLNY) return effects. Both appear as features of all three markets over the 1995 to 2010 time-frame. However, the �turn-of-the-month� effect is much more pronounced in Hong Kong and the mainland B-markets than it is in the more segmented and less international (mainland Chinese) A-market. The CLNY effect is concentrated in returns over four trading days: three days prior to and one day after the CLNY holiday. Moreover, the effect is common to all major sectors of the Hong Kong market as well as to the Shanghai and Shenzhen A- and B-markets. Despite an elevation in mean return levels at the �turn-of-the-month� and CLNY, volatility levels appear little different to other periods. In addition, as in McGuinness (200520. McGuinness , PB . 2005 . A re-examination of the holiday effect in stock returns: the case of Hong Kong . Applied Financial Economics , 15 : 1107 � 23 .
[Taylor & Francis Online] View all references), a pre-CLNY seasonal effect is absent from results. A post-CLNY seasonal effect, capturing the earnings reporting season in Hong Kong, also proved elusive. Consistent with McConnell and Xu (200819. McConnell , JJ and Xu , W . 2008 . Equity returns at the turn of the month . Financial Analysts� Journal , 64 : 49 � 64 .
[CrossRef], [Web of Science ®] View all references) for the US, we also offer no discernible evidence of a �turn-of-the-month� effect at quarter ends. Finally, and importantly, we find strong evidence that Hong Kong short-sales turnover shrinks as the calendar month-end nears. This is consistent with some participants delaying or bringing-forward short positions so as to avoid an anticipated upturn in returns at month-end.
