Empirical Legal Studies

More on Modeling Punitive Damages

In The Punitive Damages Calculus: The Differential Incidence of State Punitive Damages Reforms, Benjamin McMichael & Kip Viscusi (Vanderbilt) exploit two discrete data sets--one broad, general national sample of state civil litigation and one discrete sample of "blockbuster" civil cases which involved punitive damages awards in excess of $100M--and find that state punitive damages cap reforms influence these two streams of civil litigation differently. The paper's key findings are described below.

“This paper explores the role of punitive damages caps and other punitive damages reforms on awards using a two-part model of the judge/jury decision to award punitive damages. The punitive damages calculus involves both discrete and continuous decisions. An adjudicator must decide whether to award punitive damages as well as the amount to award. We examine the effect of state punitive damages reforms using two different data sets—the Civil Justice Survey of State Courts (CJSSC) data, which provides a broad national sample of state courts, and a sample of blockbuster punitive damages awards of at least $100 million. The most salient state punitive damages reforms—damages caps—have quite different impacts in these two samples. For the CJSSC sample, we find that punitive damages caps effectively reduce the amount of punitive damages awarded at trial by about 60% but have no statistically significant effect on the adjudicator’s decision to award punitive damages. For the blockbuster award sample, punitive damages caps lead to a decrease in the number of blockbuster awards. Thus, caps reduce the number of awards that cross the blockbuster threshold of $100 million. However, conditional on crossing that threshold, we find that caps have no statistically significant effect on blockbuster punitive damages awards.”

Truncated and Censored Data: Illustrated and Explained

A very helpful post on Stata Blog (here) discusses problems presented by two distinct (though too often conflated) phenomena that can distort samples: truncation and censoring. As the post describes: "Data are left-truncated when individuals below a threshold are not present in the sample. For example, if we want to study the size of certain fish based on the specimens captured with a net, fish smaller than the net grid won’t be present in our sample." In contrast, data are "left-censored at κ if every individual with a value below κ is present in the sample, but the actual value is unknown. This happens, for example, when we have a measuring instrument that cannot detect values below a certain level." Helpful examples in the post illustrate these issues with vivid clarity. Finally, having reminded us of a potential data problem, the post goes on to identify a potential solution. Stata commands now permit estimating the underlying population parameters for truncated data ("truncreg") as well as for censored data ("intreg" or "tobit").

President-Elect Trump's SCOTUS Nominee Pool, "Replacing" Scalia, and Measuring "Originalism"

In Searching for Justice Scalia: Measuring the "Scalia-Ness" of the Next Potential Member of the U.S. Supreme Court, Jeremy Kidd (Mercer) et al. took then-candidate Trump's promise to "appoint judges [and Justices] very much in the Justice Scalia mold," and assessed this promise against most (15/21) of those on President-Elect Trump's public "short-lists" of potential SCOTUS nominees. More specifically, the paper sets out to distill "Scalia-ness" into three vectors, develop a "Scalia Index Score," and apply the "Scalia Index Score" to 15 of Trump's possible SCOTUS nominees. While the paper, as the authors correctly note, is "not without its limitations," it nonetheless provides a glimpse into what a Trump Court might look like, at least as implied by Trump's own public statements. Obviously, the paper's analytic frame is not limited to the SCOTUS. The abstract follows.

"The President-elect promised on the campaign trail to replace Justice Scalia with a judge like Justice Scalia (and Justice Thomas). But what does it mean to be like Justice Scalia? It surely means more than just being "conservative." This study proposes three empirical measures of what made Justice Scalia Justice Scalia. First, how often does a judge promote or practice originalism? Second, how often do they cite to Justice Scalia's non-judicial writings, writings that were not about the substance of the law but about how to think about interpreting the law. And third, how often does a judge write separately, something Justice Scalia did 25.9% of the time when he was not writing the majority opinion over his last 20 years on the court. The study then applies those measures to potential nominees, and provides a metric for determining just how Scalia-like they are: the Scalia Index Score. While not without its limitations, this metric provides an objective way to evaluate how much a potential nominee is like the famous jurist they may replace."

The (Don) Songer Project--Data Access

In a fitting tribute to a leading appeals court scholar, the good folks at the Univ. of South Carolina have reorganized (and renamed) a helpful website in honor of their colleague, Don Songer. "The primary mission of the The Songer Project is to provide a comprehensive access point to the most recent and cutting-edge research on law and judicial politics. At this website, individuals interested in law and judicial politics can download electronic datasets of court cases, obtain smaller datasets or measures of judicially relevant phenomena, read various working papers on important topics, and link to other websites containing law and judicial politics information."

Well worth a visit, particularly for those seeking USCA datasets (click here).