Introduction to Special Issue on Forensic Decision Analysis

In this Special Issue, the Journal departs from its usual format in order to publish a group of interdependent papers that were written to address a major topic–the methodology for determining both the “how” and the “why” of decision-making in a litigation context. Termed forensic decision analysis by the authors, the methodology itself transcends the field of economics. However, like hypothesis testing and regression analysis, it can help economists to better answer questions of interest to them and to more fully understand the implications of alternative approaches in forensic economics. Forensic decision analysis is more a way of thinking about a problem than it is a formal theory. It involves no fixed set of assumptions or collection of formulas, though in practice its application will typically lead to numerous formulas and calculations. What is surprising, and very useful, is that the formulas and calculations often differ from those handed down to us by tradition. This message is brought into focus by the authors’ collaborative treatment of the problem of pay discrimination detection and measurement. Economists, especially labor economists, have for many years lent their expertise to courtroom proceedings, analyzing accusations of employment discrimination in terms of pay inequities. The first of the three papers in this issue is written by labor attorneys David Copus and Jaime Cole. Mr. Copus represented the Equal Employment Opportunity Commission (EEOC) in its landmark case against AT&T in the early 1970s. The EEOC employed a labor economist to formulate its case, subsequent to which AT&T agreed to a multimillion dollar settlement. The economist went on to publish the details of his analysis and, as related in this article, the context of pay discrimination litigation was set for decades to come. Mr. Copus now thinks the basis for settling that case was fundamentally flawed, and that those same flaws continue to haunt pay discrimination litigation even today. Mr. Copus and Mr. Cole outline the reasons for this discomfort, explaining that the statistical methods used by the labor economist simply do not address the questions of prime legal importance. It is left to the second and third papers in this issue to articulate an alternative to the approach that Mr. Copus and Mr. Cole decry. In the second article, David Peterson points out that many matters in litigation revolve around attempts to determine why a particular decision was made. Drawing on his experience as technical advisor to litigation teams for the past 30 and more years, Dr. Peterson illustrates in his article how this formulation applies in many types of disputes, ranging from jury selection to hiring discrimination to computer code copying, and shows furthermore how all