The sentencing guidelines were created in part to deal with a specific problem, the concern that similarly situated defendants who committed similar crimes were getting different sentences based on which judge the defendant happened to be assigned to. Under the guidelines, a judge must start a sentencing by calculating the offense level reflecting the defendant's offense, the defendant's criminal history, and a guideline range reflecting the sentence that the judge must consider in sentencing the defendant. The U.S. Supreme Court held in the 2005 U.S. v. Booker decision that the sentencing guidelines were not mandatory, but still recognized the importance of promoting uniformity in sentencing.
Unfortunately, new data from the commission shows that post-Booker federal sentencing is not turning out quite as Congress and the Supreme Court had hoped.
This most recent set of such data is contained in "Intra-City Differences in Federal Sentencing Practices," a report that was issued in January 2019. In the report, the commission looked at almost 150,000 cases where judges in 30 cities had significant discretion in sentencing and where there was sufficient information to make comparisons. Overall, the cases were about 30 percent drug cases, 14 percent illegal entry or re-entry cases, 13 percent fraud cases, 13 percent gun cases, and 29 percent other.
The report has a wealth of information and is worth studying by lawyers in the 30 cities covered. I've designed the charts below to highlight the overall trends and to make comparisons between those cities clearer.
First, the Sentencing Commission found wide differences in sentences imposed by judges within the same city, concluding: "In most cities, the length of a defendant's sentence increasingly depends on which judge in the courthouse is assigned to his or her case."
Philadelphia provides the most extreme example of the 30 cities studied by the commission. In Philadelphia, one judge on average gave sentences more than 20 percent greater than the city average, while another judge on average gave sentences around 40 percent lower than the city average. Philadelphia thus had the greatest "spread" of any major city, at 63.8. Manhattan had the second-highest spread (59.1), with two judges on average giving sentences more than 20 percent above the city average, and one judge on average giving sentences more than 30 percent below the city average.
Second, the Sentencing Commission found that most judges across the country are giving, on average, sentences well below the low end of the applicable guideline range. Out of the 30 cities studied, judges on average gave sentences below the low end of the guidelines in all but one. Average sentences in eight cities (Alexandria, Boston, Chicago, Los Angeles, Manhattan, Portland, Salt Lake City and Seattle) were more than 25 percent below the low end of the average guideline range.
Third, part of this appears to be a divide between judges regarding the weight to be given to the guidelines. Of the 339 judges whom the commission studied, there were no judges who were on average giving sentences significantly greater than the guideline minimum, but more than 90 percent of judges gave average sentences that were below the guideline minimum. In Manhattan, no judge gave sentences at or above the guideline minimum, even the ones who gave sentences far greater than their peers. Chicago only had one judge out of 24 whose average sentence was at or above the guideline minimum.
Lawyers practicing in the federal courts probably already knew much of this on an anecdotal basis, but the commission's data analysis puts some hard numbers on trends that lawyers and policymakers should be aware of and should address.
Why these differences in sentencing and these variations from the guidelines are occurring is not addressed in the commission's report. The commission does look at the mix of cases that different judges and different districts had, and this may explain some of the patterns, but probably not much, given that the analysis period covered several years' worth of sentencings.
But a look at some of the publicly available, case-specific data may provide some clues as to what is going on. Many lawyers may not be aware of this, but the commission has made a lot of anonymized data about sentencings available: first, in SAS and SPSS files that can be accessed via the commission's website, and second, in additional formats that can be accessed via the University of Michigan's Institute for Social Research.
My own analysis of the fiscal year 2016 data available online via the Institute for Social Research points to some possible factors.
First, sentences appear to go further down from the guidelines as the sentencing ranges get higher. The graph below shows that economic crimes (as indicated by Guideline 2B1.1) and drug crimes (as indicated by Guideline 2D1.1) result in sentences further below the guidelines as the ranges exceed 60 months (five years). Large cities such as New York and Chicago tend to see cases with larger loss amounts and drug quantities, and thus may see more below-guideline sentences than smaller cities with lower loss amounts and drug quantities.
Second, as seen in the graph below, many economic-crime offenders are getting below-guidelines sentences, especially as loss amounts go up. This may reflect a discount of large loss amounts, or perhaps a discounting of some of the sentencing enhancements in 2B1.1, or perhaps credit for reputational harm and loss of professional licenses.
More research and discussion is needed, but there is enough data now to make the discussion about more than just anecdotes and outlier cases. The commission's data analysis raises serious questions that should impact sentencing recommendations and perhaps help reopen a debate about balancing judicial discretion with the goal of uniformity.
First, federal prosecutors should reconsider their approach to sentencing recommendations. The U.S. Department of Justice's Justice Manual strongly suggests that prosecutors recommend sentences within the guidelines range, but also imposes a duty on the prosecutor to recommend a sentence that would not result in an unwarranted sentencing disparity.
Given the reality that judges are imposing below-guideline sentences, particularly in cases with high guideline ranges, prosecutors should consider recommending more below-guideline sentences that reflect consistency with the sentences that are actually being imposed. Prosecutors should not be making high opening bids with the expectation that the judge will split the difference. Prosecutors should make recommendations that would meet all the Section 3553 factors, including avoiding unwarranted sentencing disparities, and the reality now is that within-guidelines sentences sometimes might be unwarranted given the sentences that have been imposed post-Booker.
Prosecutors may also want to reconsider some recommendations that have been specifically tied to the low end of the guidelines. If the average sentence in a district is 20 percent or more below the guidelines, then recommending 20 percent off the low end of the guidelines for a cooperator does not mean much.
Second, defense attorneys should consider using the data in the right circumstances. Defense attorneys can look at the anonymized data available online and see how similar defendants have been sentenced, and point out cases or trends favorable to their clients. Defense attorneys in those few courtrooms where the average sentence is at or above the guideline minimum may want to point out that a within-guidelines sentence might create an unwarranted sentencing disparity given what other judges are doing.
Third, judges may want to discuss with their peers how their sentencing approaches fits with the guidelines and with Congress' intent to avoid unwarranted sentencing disparities, particularly in those districts with wide spreads in sentences.
And, fourth, policymakers and the public may want to examine the sentencing guidelines given this data. Are certain guidelines resulting in offense levels that are too high? Or does the sentencing table itself need to be recalibrated, as a one-level increase in a defendant's offense level means two or three more months at the lower levels but 30 or more months at the higher levels? And those courtrooms where sentences are particularly low might warrant some further examination.
The Booker decision properly gave judges discretion in individual cases, but the effects on overall fairness need to be studied more as many defendants' sentences appear to depend significantly on the judge that they happen to end up standing before.
Stephen Lee is a partner at Benesch Friedlander Coplan & Aronoff LLP and former assistant U.S. attorney in Chicago.
The opinions expressed are those of the author(s) and do not necessarily reflect the views of the firm, its clients, or Portfolio Media Inc., or any of its or their respective affiliates. This article is for general information purposes and is not intended to be and should not be taken as legal advice.