The Blame Game

I propose a textual analysis-based measure to detect when corporate executives blame bad performance on external factors such as industry or the economy (BLAME measure). Using this methodology to analyze quarterly earnings announcement conference call transcripts, I find that: (1) executives are more likely to blame these external factors when their companies experience bad performance, but are unwilling to credit the external factors when they perform well; (2) a high BLAME measure predicts low returns subsequent to the conference call date after controlling for the tone of the transcripts and other known predictive variables. The hedged portfolio that takes long positions in companies with low BLAME measure and short positions in companies with high BLAME measure generates abnormal returns up to 6.8% per year; (3) the BLAME measure negatively predicts earnings surprises and analyst recommendation revisions in the subsequent quarter, indicating underreaction to firm-specific negative information; (4) a high BLAME measure reduces executive turnover-performance, implying that blaming external factors reduces the punishment on executives who underperformed. Overall, the evidence suggests that investors underreact to negative information when managers attribute negative performance to external factors. ∗Emory University. E-mail: [email protected] †I am grateful for the comments and suggestions from Mengyao Cheng, Stephane Chrestien (NFA discussant), Tarun Chordia, Ilia Dichev, Clifton Green, Byoung-Hyoun Hwang, Russell Jame, Narasimhan Jegadeesh, Tim Loughran, Nadia Massoud, Joonki Noh, Breno Schmidt, Jay Shanken, Denis Sosyura, Xue Wang, Quan Wen, Steven Xiao, David Yermack and workshop participants at Emory University.