عملکرد بلند مدت سهام ها بدنبال سود تقسیمی نقدی فوق العاده و ویژه

Using essentially all declared extraordinary and special cash dividends between 1926 and 2001 which are not preceded or followed by the same for a period of three years, we find no robust post-declaration long-term abnormal stock returns, even in sub-samples defined by the special dividend yield, the bang-for-the-buck, the declaration-period abnormal return, the sub-sampling period or the stock market condition at declaration. Only event firms in the smallest CRSP market capitalization quintile display significant positive abnormal returns during the first-year following the declaration. However, these latter are not robust across sub-sampling periods. Overall, there is no compelling evidence that investors under- or over-react to extraordinary or special cash dividends.

We begin our analyses by examining the post-declaration long-term abnormal stock returns for the entire sample. Table 2 reports the results for the first, second and third post-declaration years and for the one-, two-, and three-year post-declaration periods. The equally-weighted average abnormal monthly returns during the first-year following the declaration date, using the OLS and the WLS methods, are positive and their magnitude is close 0.18 and 0.16%, respectively. These monthly abnormal returns translate to only 2.18 and 1.93% when compounded over the one year, which is an economically small abnormal performance. In any case, while the WLS estimate of 0.16% is statistically significant at the 5% level, the OLS estimate of 0.18% has an unimpressive t-statistic of 1.43. Also, when portfolios are value-weighted by market capitalization, the average abnormal monthly returns during the first-year following the declaration date are much smaller in magnitude (0.02%), and the WLS estimate has a negative sign. Clearly, neither the magnitude, nor the sign, nor the statistical significance of the first-year post-declaration abnormal return is robust to changing the weighting method or the estimation procedure. In addition, all the abnormal returns over the second and third post-declaration years and over the two- and three-year post-declaration periods are statistically insignificant. Therefore, there is no clear evidence that investors generally under- or over-react to declarations of extraordinary or special U.S. cash dividends. We examine next whether our aggregate/general results for the entire sample mask some pricing inefficiency(ies) in some sub-samples.