Assessing punitive damages (with notes on cognition and valuation in law)

Although legal scholars have disagreed about whether juries should be allowed to award punitive damages and about how judges should instruct them, the debate has included little discussion of jurors' cognitive capabilities. In this Article, Professors Sunstein, Kahneman, and Schkade respond to this gap by offering an experimental study. The study seeks to separate the tasks that a jury is suited to perform from those that a jury can accomplish only with great inconsistency. In personal injury cases, the study shows, jurors' normative judgments about outrageousness and appropriate punishment are relatively uniform, at least when measured on a bounded numerical scale (0 to 6). Indeed, these normative judgments are uniform across race, age, education, wealth, and gender When subjects map their judgments onto an unbounded dollar scale, however outcomes become erratic and unpredictable. Drawing on these results, the authors question the current legal approaches to the regulation of punitive damages. They then analyze various reform proposals designed to overcome erratic awards, including damage caps, compensatory judgement "multipliers," and conversion formulas that translate either jury judgments on bounded numerical scabs or jury arrangement of comparison cases into punitive damage awards. Finally the authors discuss the implications of the study for many other issues of law: including contingent valuation and compensatory damages in such areas as pain and suffering, libel, sexual harassment, and intentional infliction of emotional distress.