Transactional Records Access Clearinghouse (“TRAC”) has a new report titled Equal Justice and Sentencing Practices Among Federal District Court Judges, here. The email I received with a summary of the report was titled: “The Impact of the Identity of the Judge on Sentencing.”
TRAC gathers and maintains a lot of data and ways to access and analyze data at its web site, here. For example, TRAC provides reports and bulletins on the IRS . Scholars, practitioners, and students should familiarize themselves generally with the TRAC offerings and studies.
The particular TRAC offering discussed here on sentencing practices should be interesting for scholars, practitioners and students of federal tax crimes. Sentencing, after all, is where the rubber hits the road so to speak. So, I offer some excerpts first from the email summary and then the report (with some redundancy):
Excerpts from the Email Summary
While judges need sufficient discretion to consider the totality of circumstances in assigning a sentence in a specific case to ensure it is “just,” a fair court system always seeks to provide equal justice under the law, working to ensure that sentencing patterns of judges are not widely different for similar kinds of cases.
While special circumstances might account for some of these differences, half of the courthouses in the country had median differences in prison sentences of 16 months or more, and average differences of 21 months or more. Five courthouses showed more than 60 months difference in the median prison sentence handed out across judges serving on the same bench.
In contrast, seven federal courthouses out of 159 compared had perfect agreement among judges in the typical or median sentences assigned. In an additional thirty, judge-to-judge sentences differed by six months or less. This study largely replicates the findings from TRAC’s first national judge-by-judge examination of the differences among federal judges in sentencing practices that was published almost a decade ago. TRAC compared judge decisions within each courthouse to take advantage of the broad institutional promise that criminal cases there are assigned to judges on a random basis. Just as this methodology based upon random assignment is the gold standard for determining whether a new vaccine or the adoption of new medical procedures have any real impact on patient outcomes, in a similar fashion this methodology provides the most reliable method for isolating the influence of a particular judge’s sentencing philosophy from the impact of the multitude of other factors impacting sentence outcome.
Excerpts from the Report (footnotes, figures and tables Omitted)
This report examines very recent data on federal trial judges and their sentencing practices. The existence of judge-to-judge differences in sentences of course is not synonymous with finding unwarranted sentencing disparity. Most people agree that a key requirement for achieving justice in this country is that the judges in a court system have sufficient discretion to consider the totality of circumstances in deciding that a sentence in a specific case is just. No set of rules, including the Federal Sentencing Guidelines, can substitute for this necessary individualized judgment.
But a fair court system always seeks to provide equal justice under the law, working to ensure that sentencing patterns of judges not be widely different when they are handling similar kinds of cases.
In reality, sometimes the goal of equal justice under the law is achieved, and other times the actual sentences handed down depart markedly from this goal. Using case-by- case, judge-by-judge, data updated through December 2020, a new analysis by the Transactional Records Access Clearinghouse (TRAC) at Syracuse University identifies federal courthouses where wide judge-to-judge sentencing differences currently occur, and courthouses where there is wide agreement in sentencing among judges.
While special circumstances might account for some of these differences, half of the courthouses in the country had median differences in prison sentences of 16 months or more, and average differences of 21 months or more.
Results further showed that currently seven (7) federal courthouses out of 159 compared had perfect agreement among judges in the typical or median sentences assigned. In an additional thirty (30), judge-to-judge sentences differed by six months or less. The seven courthouses with perfect agreement included those at; Lincoln, Nebraska; Providence, Rhode Island; Albany, Georgia; Ft. Myers, Florida; Las Cruces, New Mexico; and El Paso and Del Rio, Texas.
At the other extreme, five (5) courthouses showed more than 60 months difference in the median prison sentence handed out across judges serving on the same bench. These five courthouses with extreme differences were: Tampa, Florida; Benton, Illinois; Orlando, Florida; Greenbelt, Maryland; and Philadelphia, Pennsylvania.
* * *
TRAC’s Study and Random Assignment
It would be inappropriate to compare sentences handed down by judges sitting in different courthouses. Case characteristics vary widely among communities in the country. This means that just because two judges hand down typical sentences of different lengths, it is plausible that these differences simply reflect actual differences in the nature, statutory penalties, and seriousness of the crimes in each judge’s caseload rather than real differences in their sentencing philosophies. Even comparisons which focus on sentencing practices for particular classes of offenses, such as drugs or white-collar crimes, still are potentially left with substantial variation in the nature and seriousness of the makeup of offenses within each of these broad categories.
Instead, TRAC compares judges within the same courthouse. This approach takes advantage of the broad institutional promise that criminal cases within the same courthouse are assigned to judges on a random basis. When random assignment occurs, we know by the law of large numbers that—assuming a reasonably large number of cases—the composition of the cases each judge hears will be roughly comparable. Just as this methodology based upon random assignment to different groups is the gold standard for determining whether a new vaccine or the adoption of new medical procedures have any real impact on patient outcomes, in a similar fashion this methodology provides the most reliable method for isolating the influence of a particular judge’s sentencing practices from the impact of the multitude of other factors impacting sentence outcome, including the nature and seriousness of the crime and the defendant’s personal history.
To ensure a sufficient number of cases, we limited our analysis to federal district court judges who sentenced at least 50 defendants at a given courthouse. A total of 159 courthouses had at least two judges who had a sufficient volume of cases to compare. From January 2016 through December 2020, 873 federal district court judges met this standard. Together, they sentenced a total of 301,326 defendants. The median prison sentence in months for each judge was calculated and then compared with that for the remaining judges serving on the bench at the same courthouse.
The median sentence was used for comparison since it represents a typical sentence for that judge with half of defendants given a longer sentence and half shorter terms. Unlike the average sentence, the median sentence is not influenced at all by extremely long sentences that could occur in unusual cases.
Individual reports by name on each judge have just been published and are available at TRAC’s Judge Information Center. Each report compares that judge’s prison sentences to each other judge at that courthouse, and provides details on the composition of each judge’s caseload as compared with others to see if they were in fact comparable.
1. I am sure all are familiar with the quip, variously attributed, that there are “lies, damned lies and statistics.” See Wikipedia entry, “Lies, damned lies, and statistics,” here. There is another aphorism which cuts the other way: “It is easy to lie with statistics, but easier to lie without them,” attributed to Fred Mosteller, one of the most eminent statisticians of the 20th Century. Wikiquote entry, “Statistics,” here (viewed 2/20/19).
2. With the caveat about statistics, I assume that the Sentencing Commission keeps similar data to that offered by TRAC in this report. The Sentencing Commission website offers a data analyzer here. The data analyzer, however, does not offer (i) the data by sentencing characteristics (including the Guidelines factors) or (ii) the data by Judge. The point of the TRAC data analysis is that, in some cases, it matters who is the judge. Observers of the sentencing scene have known that the judge matters for as long as they observed.
3. Even more focused on tax crimes, I assume that DOJ Tax CES keeps such data with even more analysis points (or fields in database-speak) than the Sentencing Commission. The Sentencing Commission may have more analysis points (or fields) than it offers on the web. For example, the sentencing judge is likely such a field that both the Sentencing Commission and DOJ Tax CES should compile. (If they don’t, they are dropping the ball; perhaps a well-targeted FOIA request to DOJ Tax might get the data or at least some acknowledgment as to whether it exists.)
4. While I don’t have the data to support any conclusions, I can make certain conclusions with the caveat that it is necessarily anecdotal from my limited observation of the field:
· White collar crimes usually are sentenced more leniently (measured by comparison to the Guidelines ranges) than non-white collar crimes.
· Within the white collar crimes set, tax crimes are sentenced even more leniently (suggesting, for example, that stealing money from the Government is more socially acceptable than stealing money from a citizen in a white collar crime context (such as embezzlement)).
I can’t back that up with statistics. I think I read an article recently consistent with these conclusions, but cannot put my hands on it right now. If I find it, I will update this blog entry.