US Sentencing Commission releases mid-FY 2018 sentencing data with re-engineered accounting of departures and variances

US Sentencing Commission has now released here its latest quarterly data report, and this one "contains preliminary quarterly data on cases in which the offender was sentenced during the first half of fiscal year 2018" (which is the period from October 1, 2017, through March 31, 2018).  I do not believe the USSC released first quarter FY 2018 data, so this new report seems to be the first big data report of the "post-Sessions Memo era" -- i.e., since AG Jeff Sessions issued his May 2017 charging and sentencing memorandum directing federal prosecutors to pursue those offenses that carry the most substantial guidelines sentence, including mandatory minimum sentences, and to more regularly seek within-guideline sentences.

From a quick glance and comparing this data from the last full year of sentencing data from the Obama Administration (in this FY 2016 data report), there does seem to be a noticeable uptick in mean sentences in some big crime categories.  For example (drawing from Table 6 in both data runs): the mean drug trafficking sentence was 75 months in the first half of FY 2018, the mean in FY 2016 was only 66 months; the mean fraud sentence was 27 months in the first half of FY 2018, the mean in FY 2016 was only 25 months.  But, interestingly, in other big crime categories there was a downtick in mean sentences: the mean firearm sentence was 70 months in the first half of FY 2018, the mean in FY 2016 was 75 months; the mean immigration sentence was 11 months in the first half of FY 2018, the mean in FY 2016 was 13 months.  Putt this all together with other less common offenses, and it turns out the cumulative mean federal sentence for the first half of FY 2018 was 45 months, the exact same mean for all federal sentences in FY 2016.

I would report some similar comparable data on departures and variances, but the USSC in this data run has significant altered how it accounts and reports this data.  Here is part of the USSC's explanation of its new accounting:

Beginning with this report, the Commission is again updating the way it presents quarterly data.  In this report, all analyses that involve a comparison of the position of the sentence imposed to the guideline range that applied in the case are presented in a new way.  Sentences are now grouped into two broad categories: Sentences Under the Guidelines Manual and Variances.  The former category comprises all cases in which the sentence imposed was within the applicable guideline range or, if outside the range, where the court indicated that one or more of the departure reasons in the Commission’s Guidelines Manual was a basis for the sentence.  Variance cases are those where the sentence was outside the guideline range (either above or below) and where the court did not cite any guideline reason for the sentence.  Data for important subgroups within these two categories are also reported.

In other words, within-guideline and "traditional departure" sentences are grouped together, while all Booker-allowed variances broken out distinctly.  It seems that all the key data previously reported on Table 8 of past USSC's data reports still appears in Table 8A of the new report.  But, fascinatingly, the new organization showcases now that roughly 3/4 of all sentences (74.7% to be exact) are "Sentences under the Guidelines Manual" with "variances" now accounting for only 25.3% of the sentences (with 2% being upward variances, 5.5% being "government motion" variances and 17.7% being "non-government" variances). 

Repackaging aside, we can still look at the "within-guideline" number on Table 8 and 8A for direct comparisons on this front between the first half of FY 2018 and all federal sentences in FY 2016.  Doing so shows that the within-guideline sentencing rate has increased from 48.6% in FY 2016 up to 50% in the first half of FY 2018.  Without a more intricate and sophisticated analysis controlling for caseloads and other factors, it is too hard to say there is conclusive evidence that the Sessions Memo is having a real impact on federal sentencing outcome.  But these data are suggestive of trends that seem likely to continue as move cases more through the pipeline and as a new set of federal prosecutors give effect to commands from Main Justice.