Friday, September 21, 2012

Behavioral Bias-The Hidden Risk in Mergers and Acquisitions

 Mergers and acquisition decisions are made by humans, not mathematical models. Humans have biases and make mistakes when fear or greed overrides their analysis. Richard Roll noticed this in his 1986 hubris hypothesis. These systematic deviations from rational calculations lead to substantial value destruction. As Warren Buffett notes, any CEO business craving will be quickly supported by detailed financial and strategic studies. Thus, it is better to start our review at the top with the CEO. The issues are understandable given the infrequent nature of acquisitions, limited feedback and the delay between upfront benefits and later costs. The errors can be both failing to make the good acquisition and making the wrong acquisition. Our focus here is on the latter.

Some major senior management biases are as follows:

1) Herding: mindlessly imitating peer actions - the “I have to have one because all the other kids have one” excuse. It gets especially bad as the merger cycle peaks resulting in buying at a high price.

2) Anchoring: overweighing certain information such as the original bid price or locking on to the 52 week high of the target’s shares for the opening bid.

3) Over Optimism and its cousin Over Confidence: this is the “we can do it” attitude to justify any synergy estimate needed to defend the price.

4) Confirmation: over weighting confirming evidence to the exclusion of contradictory facts.  These lead to a win at any cost approach known as the winner’s curse in which the “winner” overpays.  A practical way to handle the winner’s curse problem is to set a maximum price before the bidding starts.  Attempts to increase the price after the bidding starts should be questioned.
Remember we are often:

  •       Guided by selective memories 
  •     Often fail to consider what we believe to be false
  •     Influenced by the actions of others
  •     Confusing preferences with predictions
  •     Engaging in self serving attribution
  •     Denigrating non conforming views
Hopefully, forewarned is forearmed.


No comments:

Post a Comment