Accelerated share repurchase (ASR) programs, an innovative way of repurchasing shares, have become increasingly popular in recent years. In this paper, we analyze firms' rationale for undertaking ASRs rather than the traditional open market repurchase (OMR) programs. Using a hand-collected sample of ASR announcements, we test eight hypotheses regarding firms' rationale for conducting ASR programs: the distribution of excess cash hypothesis, the target leverage-ratio hypothesis, the takeover avoidance hypothesis, the employee stock option dilution hypothesis, the managerial opportunism hypothesis, the liquidity reduction hypothesis, the EPS manipulation hypothesis, and the signaling or undervaluation hypothesis. We find that firms undertaking ASR programs are significantly larger than those undertaking OMR programs, and that ASR programs have a larger median dollar amount of deal values than OMR programs. Further, ASR firms have significantly smaller cash holdings, higher dividend payout ratios, higher pre-announcement industry-adjusted leverage ratios, and similar probabilities of being takeover targets compared to OMR firms, and ASR firms grant fewer stock options (scaled by sales) to their employees than OMR firms. Option exercise by executives does not increase following buyback announcements in either ASRs or OMRs. Stock liquidity increases following both ASRs and OMRs. Although our univariate tests reveal that ASR firms are more likely to tie the CEO bonus to EPS, our multivariate analysis does not find that EPS manipulation is a significant factor in firms choosing an ASR over an OMR program. Finally, firms undertaking ASR programs have lower pre-announcement market valuations, greater positive announcement effects, and better post-announcement operating and stock return performance, compared to those announcing OMR programs. Overall, our results are consistent with the predictions of the signaling/undervaluation hypothesis but inconsistent with those of the other seven hypotheses
