We find that board gender diversity increases the likelihood that firms announce a buyback but long-term excess returns are significantly smaller when there is larger female representation on the board. This is consistent with the governance hypothesis: gender diversity makes it more likely that firms buy back stock to reduce agency costs of free cash flow. But because gender diversity improves the quality of public information disclosure repurchases are less driven by market timing. Moreover, when the quality of monitoring is lower because board members sit on many other boards, long-term excess returns are larger.

The buyback anomaly survives when using the five factor Fama and French (2015) and the four factor Stambaugh and Yuan (2017) models: buyback announcements are followed by positive long-term excess returns that are positively related to (idiosyncratic) volatility, inconsistent with the low volatility anomaly. The results are consistent with the costly arbitrage hypothesis (Stambaugh et al., 2015) as well as with the market timing hypothesis: the option to take advantage of undervalued stock is more valuable when firm value is more uncertain or is more driven by company-specific information. Combining volatility with undervaluation indicators proposed by Peyer and Vermaelen (2009) improves the predictability of excess returns after buyback announcements.

We develop an approach that combines the estimation of monthly firm-level expected returns with an assignment of firms to (possibly) latent groups, both based on observable characteristics, using machine learning principles with linear models. The best-performing methods are flexible two-stage sparse models that capture group-membership predictive relationships. Portfolios formed to exploit such group-varying predictions based on a parsimonious set of characteristics deliver economically meaningful returns with low turnover. We propose statistical tests based on nonparametric bootstrapping for our results, and detail how different characteristics may matter for different groups of firms, making comparisons to the existing literature.