Limits of arbitrage and mispricing : evidence from mergers and acquisitions

Purpose: This paper examines the extent to which noise demand and limits of arbitrage affect the pricing of acquirer stocks both at the announcement period and over the longer horizon. Design/methodology/approach: An event study approach was adopted to measure announcement-period cumulative abnormal returns. Long-horizon returns are measured using buy-and-hold abnormal returns (BHARs), calendar time portfolios (CTPRs), and subsequent earnings announcement period abnormal returns. Main methodologies include ordinary least squared (OLS) regressions, Logit regressions, and portfolio analysis. Findings: (1) Acquirer stocks with high idiosyncratic volatility (the proxy for the security level characteristic most directly associated with limits to arbitrage) earn higher announcement-period abnormal returns. (2) The return pattern reverses over the subsequent longer horizon, resembling news-driven transitory mispricing. (3) The mispricing is greater when deal and firm characteristics exacerbate the limits of arbitrage, and it weakens over time. (4) Transactions by higher idiosyncratic volatility acquirers are more likely to fail. Originality/value: Limits of arbitrage theory have been tested mostly in information-free circumstances. The findings in this paper extend the supporting evidence for limits of arbitrage explaining mispricing beyond the boundaries of information-free circumstances.