In their recent Arizona Law Review article entitled What Juries Can't Do Well: The Jury's Performance As a Risk Manager,' Professors Reid Hastie and W. Kip Viscusi purport to show that juries are likely to do a poor job in setting punitive damages, largely because jurors cannot avoid the influence of what is called "hindsight bias," or the tendency to see the likelihood of an event higher in retrospect than it would have appeared before it happened. In particular, they argue that hindsight bias and other cognitive biases undermine the utility of jury-set punitive damage awards as risk management devices. They also present information which they think shows that judges do a better job than juries in setting punitive damage levels. Their claims are interesting, but they are either wrong or unsupported by the evidence they present. In particular, if punitive damage awards are not a useful risk management device, jury hindsight bias, as I shall show, is not an important reason.
Lempert, Richard O. "Juries, Hindsight, and Punitive Damage Awards: Failures of a Social Science Case for Change." DePaul Law Review 48, no. 4 (1999): 867-894.