Errors in Estimating Patent and Class Action Damages
57 Pages Posted: 24 Aug 2019 Last revised: 30 Aug 2019
Date Written: August 21, 2019
How much are consumers willing to pay for the rounded edges of an iPhone? How much more are consumers willing to pay for a soap that kills 99% of bacteria as opposed to 90%? Answering these questions is at the heart of estimating damage awards for patent infringement and false advertising class action lawsuits respectively. But these questions are difficult and often require the use of social science experts. These experts use complicated empirical black box methodologies to introduce opinions leading to billions of dollars of damages.
One such method that has gained much traction in federal courts is “choice based conjoint.” This method seeks to answer how much consumers are willing to pay for various features of products. Although a powerful and widely accepted method outside of the law, it is currently being misused in its application to patent infringement and false advertising lawsuits. Experts have been applying the method incorrectly and judges, lawyers, and litigants are not adequately policing the use of the method.
In this article, we explain how the method applies to the legal context and open up the proverbial black box so that legal practitioners can be savvy consumers of the method. More importantly, we empirically show through our novel experimental design and using Hierarchical Bayes estimation, how the method often has been inappropriately applied in the legal context and used to validate unjust, excessive, and unrealistic damage awards.
Keywords: patent, class action, consumer protection, marketing, empirical, law
Suggested Citation: Suggested Citation