Empirical Legal Studies

AAAS & Ted Eisenberg

It is with great pleasure that I note that my friend, frequent coauthor, and Cornell colleague--Theodore Eisenberg--is among the eight law professors recently elected to the American Academy of Arts and Sciences. A well-deserved and long overdue honor. The complete list is here.

Sentencing Studies and Quasi-Natural Experiments

Today's New York Times reports that one of California's responses to its crushing fiscal mess is to implement a new (and much more aggressive) early release program for non-violent prisoners who are unlikely to recidivate. As Laura Appleman notes over at The Faculty Lounge, California's new early release program is being closely watched by other cash-strapped states also looking to reduce spending. Assuming that at least some states do not experiment with dramatic changes to early release programs, however, California's program might generate helpful data for those seeking to test the efficacy of its new policy.

Legal Academic Hiring Market, JD/PhD Debate, and the "Empirical Revolution"

As usual, Dave Hoffman (Temple) raises some quite interesting points over at Concurring Opinions (and here).

All You Need to Know About H.R. 3962

The good folks over at the Computational Legal Studies blog have an interesting take on the recently-passed H.R. 3962 (health care). Befitting the blog's overall "computational or complex systems" approach, the post contains more than a few interesting factoids. As Dan Martin Katz notes, "The bill is 1990 pages as 'typeset' by the House.  However, the 'substantive' words contained in bill are no longer than the length of a Harry Potter Novel."

Atkins' Influence on Death Row Litigation

The Supreme Court's 2002 Atkins decision raised interesting questions amenable to empirical analysis. In An Empirical Look at Atkins vs. Virginia and its Application in Capital Cases, Cornell colleagues (and leading death penalty defense attorneys) John Blume, Sheri Johnson, and Christopher Seeds, publish preliminary results from their look-back study of Atkins' influence. The abstract follows.

"In Atkins vs. Virginia, the Supreme Court declared that evolving standards of decency and the Eighth Amendment prohibit the death penalty for individuals with intellectual disability (formerly, "mental retardation"). Both supporters and opponents of the categorical exemption, however, have criticized the Atkins opinion. The Atkins dissent, for example, urged that the decision would open the gates of litigation to a flood of frivolous claims. Another prominent criticism, heard from those more supportive of the Court's ruling, has been that the language the Court used—communicating that states must 'generally conform' to the clinical definitions of mental retardation—is ambiguous enough to permit states to stray from the clinical definitions and, consequently, for death eligibility to vary depending upon the jurisdiction in which a defendant is charged. This Article presents preliminary data responsive to these issues, and reports three basic findings. First, Atkins has not opened floodgates of non-meritorious litigation. Second, the success rates for Atkins claims vary dramatically between states and state deviations from the clinical definitions appear to have a palpable impact. Third, as compared to their representation on death row, African-American defendants both file and win a disproportionately high number of Atkins claims."

"Empirical Holy War" -- Eisenberg v. US Chamber of Commerce

Reaction to a recent post on the Wall Street Journal Law Blog (here) makes it clear that my colleague Ted Eisenberg's (Cornell) data-driven critique (forthcoming in JELS) of the US Chamber of Commerce's annual state "rankings" of perceived "lawsuit climate" struck something of a raw nerve. As the WSJ notes, "The Chamber’s Institute for Legal Reform publishes an annual survey,  conducted by the Harris Poll, which ranks states from best to worst, according to how their judges treat class actions, the prevalence of punitive damages, and juries’ “predictability and fairness” among other factors." The WSJ's brief summary of Ted's critique of the survey follows:

"Eisenberg concludes that the survey itself is logically inconsistent. Respondents, for example, he writes, make incorrect conclusion about which states are the most favorable towards punitive damages due to erroneous impressions among the respondents or even misrepresentations of a state’s laws by the Chamber. He says that a state’s movement in the ranking — such as New Jersey’s, which plummeted in one year — could not be validated by a detailed analysis of legal events in the state. Another problem: respondents are given the prior year’s rankings before offering their opinions. This reduces the likelihood that a state could shake an unfavorable image, even if state law there had changed."

Anyone Using Stata v.11 Yet?

Before I finally commit to upgrading to Stata v.11 I'd appreciate any comments (pro or con) from anyone already using it. Major new versions of any software scare me a bit.

The End of an Era: the Bi-Modal Distribution for the Class of 2008

NALP has just posted its entry-level starting salary for class of 2008--i.e., the lawyers who started their jobs just as Bear Sterns and Lehman Bros unraveled and the credit markets completely froze up. 
 
Of the 22,305 law school graduates in NALP's sample (over half of all 2008 graduates), a remarkable 23% (5,130  '08 grads) reported an entry-level salary of $160,000.  In contrast, 42% of entry level lawyers reported salaries in the $40,000 to $65,000 range.  Once again, the central tendencies are a poor guide to the distribution as a whole: whereas the mean salary is a $92,000, the median salary was $72,000.   Further, the two modes ($50,000 and $160,000) are separated by $110,000.

Amidst all the layoffs, deferrals, salary cuts, and apprenticeship programs announced in 2009, it is safe to venture that the bi-modal era has peaked.  Every law school class for the foreseeable future will graduate to a much different economic landscape.  Although many students will regret the opportunity to earn such a big payday upon graduation, it brought with it intense billing pressure, client resentment, heavy leverage, and very little substantive training for new hires.  I would argue that profession as a whole (including current and future graduating classes) is better off with a lower entry level salary. 

Admittedly that is a long-term view for the profession as a whole.  In the short term, current students and recent graduates are in a world of hurt.  Specifically, law school debt loads continue to climb.  Thus, law schools are (rightfully) going to be under increased pressure to deliver value to our students.  I don't think most law professors and law school administrators fully appreciate the difficult times ahead.  For a provocative take on the current state of legal education, see Paul Lippe, Welcome to the Future: Time for Law School 4.0.

For some perspective on how this crazy market evolved, see:

• Henderson, Distribution of Starting Salaries for 2006: Best Graphic of the Year (Sept. 4, 2007)
• Henderson, How the Cravath System Created the Bi-Modal Distribution (July 18, 2008)
• Henderson, Class of 2007:  A More Extreme Bi-Modal Distribution (July 30, 2008)
• NALP, A Picture is Worth 1,000 Words (Sept. 2007).

After the jump are the distributions from 1991, 2006, and 2007.  The primary takeaway is that the bi-modal did not exist in the early 1990s.  It first emerged in 2000 (with the dot.com salary wars) and became progressively more extreme starting as the decade unfolded.  On Wednesday, I have an article coming out in the NALP Bulletin, entitled "The Bursting of the Pedigree Bubble," which will provide some additional analysis.

Supremely Empirical

As long as every other interested party is weighing in on Pres. Obama's first S.Ct. nomination, the least I can do is endorse Dave Hoffman's proposal for a nominee that, at a minimum, possesses a general familiarity and comfort with empirical methods. Although I can understand it when judges choose to simply ignore salient (and, I hope, well-crafted) empirical evidence brought before them (however much it might pain me), when judges (especially Justices) choose to engage with empirical work, however, they really should avoid botching things. In Exxon, Justice Souter's recent high-profile bungling (or "debacle" as Dave notes here; subsequent scholarly work on the issue is found here) falls below what I expect from my law students.  Dave's conclusion warrants repetition:

"The point is that at least some exposure in statistics and social science techniques is quickly becoming part of a well-rounded legal education. It should also be part of what we look for in a Justice."

Judicial Pay Debate Data

Over at the Volokh Conspiracy Eric Posner brings logitudinal data to the judicial pay debate. The data come from a paper on the same topic, Are Judges Overpaid?: A Skeptical Response to the Judicial Salary Debate, by Stephen Choi (NYU), Mitu Gulati (Duke), and Eric Posner (Chicago).