Sometimes you just get lucky but overconfident CEOs never talk about good luck when things go well, just bad luck when there are problems, according to a paper in the current issue of Management Science.

Whether to engage in mergers and acquisitions is one of the most important decisions top managers make, the authors write. While many of the factors influencing these decisions may be based on objective financial metrics, there is increasing evidence that behavioral biases play an important role in managerial decision making.

Professors Matthew T. Billett and Yiming Qian of the University of Iowa based their results on a sample of public acquisitions between 1985 and 2002. Over this period, U.S. public companies acquired $3.7 trillion worth of other U.S. public companies.

The authors explore one such bias -- managerial overconfidence -- and find evidence suggesting CEOs develop overconfidence through 'self-attribution bias' when making merger and acquisition decisions. Individuals subject to self-attribution bias overcredit their role in bringing about good outcomes and underestimate the role of luck.

Consistent with this, they find that CEOs appear to overly attribute their role in successful deals, leading to more deals even though these subsequent deals are value destructive.

They also find evidence that CEOs alter their stock holdings prior to deals in a pattern consistent with overconfidence in the outcome of these subsequent deals.

The authors advise that CEOs be particularly cautious and disciplined when engaging in acquisitions following prior success. Boards and other stakeholders should also ensure that any proposed deal is judged on its own merits and is not justified on the basis of prior CEO success in mergers and acquisitions, they say.

Article: Matthew T. Billett, Yiming Qian,'Are Overconfident CEOs Born or Made? Evidence of Self-Attribution Bias from Frequent Acquirers', MANAGEMENT SCIENCE Vol. 54, No. 6, June 2008, pp. 1037-1051 DOI: 10.1287/mnsc.1070.0830