Honda Motor Co. v. Oberg, 512 U.S. 415, 19 (1994)

Page:   Index   Previous  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  Next

Cite as: 512 U. S. 415 (1994)

Opinion of the Court

limitation of punitive damages to the amount specified, is hardly a constraint at all, because there is no limit to the amount the plaintiff can request, and it is unclear whether an award exceeding the amount requested could be set aside. See Tenold v. Weyerhaeuser Co., 127 Ore. App. 511, 873 P. 2d 413 (1994) (Oregon Constitution bars court from examining jury award to ensure compliance with $500,000 statutory limit on noneconomic damages). The second safeguard, the clear and convincing standard of proof, is an important check against unwarranted imposition of punitive damages, but, like the "no substantial evidence" review discussed supra, at 429, it provides no assurance that those whose conduct is sanctionable by punitive damages are not subjected to punitive damages of arbitrary amounts. Regarding the third purported constraint, respondent cites no cases to support the idea that Oregon courts do or can set maximum punitive damages awards in advance of the verdict. Nor are we aware of any court which implements that procedure. Respondent's final safeguard, proper jury instruction, is a well-established and, of course, important check against excessive awards. The problem that concerns us, however, is the possibility that a jury will not follow those instructions and may return a lawless, biased, or arbitrary verdict.11

11 Respondent also argues that empirical evidence supports the effectiveness of these safeguards. It points to the analysis of an amicus showing that the average punitive damages award in a products liability case in Oregon is less than the national average. Brief for Trial Lawyers for Public Justice as Amicus Curiae. While we welcome respondent's introduction of empirical evidence on the effectiveness of Oregon's legal rules, its statistics are undermined by the fact that the Oregon average is computed from only two punitive damages awards. It is well known that one cannot draw valid statistical inferences from such a small number of observations.

Empirical evidence, in fact, supports the importance of judicial review of the size of punitive damages awards. The most exhaustive study of punitive damages establishes that over half of punitive damages awards were appealed, and that more than half of those appealed resulted in reductions or reversals of the punitive damages. In over 10% of the cases

433

Page:   Index   Previous  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  Next

Last modified: October 4, 2007