Tales of an Empirical Nothing (Part 1): Measuring the Impact of Judicial Nominations on the Ideology of Courts of Appeals

As a practicing lawyer in a big firm one of your expected acts of "service" (and business development) is to speak at seminars.  So, as an academically inclined labor and employment lawyer at Littler, I started doing an annual presentation for the North Carolina and South Carolina Bar Associations on the labor and employment law decisions of U.S. Court of Appeals for the Fourth Circuit. 

As most court watchers know, the Fourth Circuit has, historically, been one of the most conservative of the federal Courts of Appeals.  However, about 5 years ago, I began to notice a moderation in some of the Fourth Circuit's labor and employment decisions.  This perceived moderation, of course, seemed to correspond with the confirmation of President Obama's first nominees to the Fourth Circuit bench and seemed to increase as more of his nominees were confirmed. 

Some of the Fourth Circuit's high-profile cases over the last few years have seemed very moderate (if not liberal).  [See, e.g., Bostic v. Shaefer (4th Cir. 2014)(declaring Virginia's ban on same sex marriage unconstitutional); King v. Burwell (4th Cir. 2014)(finding ACA tax subsidies valid); and Liberty Univ. v. Geithner (4th Cir. 2011)(holding that penalties for non-compliance with the ACA's individual mandate constituted a “tax”).  However, given the 20+ years of solidly conservative circuit precedent the court had built up, I wondered if a true ideological shift was possible and whether the moderation I perceived reflected a true shift in the Fourth Circuit's ideology or if it was the result of the individual cases, of panel effects, or of something else. 

So, I set out to try to prove (a) whether a "leftward" ideological shift was occurring or had occurred on the Fourth Circuit and (b) whether the influx of President Obama's successful nominees caused it. 

To prove these things, I, a former litigator whose highest achievement in mathematics was heretofore a B-minus in Trigonometry in college, had to learn something of study design, case coding, and -- most terrifyingly -- statistics. 

On the first two, I learned through trial and error and eventually realized that I had essentially set out to reinvent the wheel.  Instead of using preexisting databases and coding, I created my own dataset and my own coding scheme.  Rookie mistake.  Oh well.  At the end of the day -- or year -- I had a dataset of all of the 4th Circuit's labor and employment law decisions from the years 2004, 2006, 2008, 2010, and 2012; coded by outcome.  I presented some very rudimentary data analysis on these years at the Seventh Annual Colloquium on Scholarship in Employment & Labor Law. 

Based on comments there (particularly from Charlie Sullivan [Seton Hall]), it was clear that I needed a "control" group to compare to the 4th Cir., which was my study group.  So, I looked for the Court of Appeals on which Pres. Obama had the least influence during his first term.   The Eighth Circuit, to which Pres. Obama had no nominees during his first term, fit the bill.  With my control group identified, I created a dataset for the Eighth Circuit that was the same as the one for the Fourth Circuit: all labor and employment decisions from 2004, 2006, 2008, 2010, and 2012, all of which were coded by outcome.

So, now I had my dataset consisting of 388 opinions from the Fourth Circuit and 418 opinions from the Eighth Circuit, coded by outcomes.  The total outcomes looked different.  But, to know if the differences were meaningful, I had to learn about statistics. 

TO BE CONTINUED ...

Comments

[Derek Tokaz]

Minor (actually major) quibble: You probably won't be able to 'prove' whether Obama's nominees caused the shift.

Best case scenario, pre-Obama judges don't change their voting patterns, Obama's judges are the deciding votes, and the 8th Circuit doesn't change either.

That makes for a very compelling case, but proof of causation is very tricky.