« Latest update on battles over criminal justice reform inside the Trump Administration | Main | "What If Prosecutors Wanted to Keep People Out of Prison?" »

March 28, 2018

"Recidivism Among Federal Offenders Receiving Retroactive Sentence Reductions: The 2011 Fair Sentencing Act Guideline Amendment"

The title of this post is the title of this notable and timely new report from the US Sentencing Commission. Here is a summary of its coverage and findings from this USSC webpage:

The publication Recidivism Among Federal Offenders Receiving Retroactive Sentence Reductions: The 2011 Fair Sentencing Act Guideline Amendment analyzes recidivism among crack cocaine offenders who were released immediately before and after implementation of the 2011 Fair Sentencing Act Guideline Amendment, and followed in the community for three years.

In order to study the impact of retroactive sentence reductions on recidivism rates, staff analyzed the recidivism rate for a group of crack cocaine offenders whose sentences were reduced pursuant to retroactive application of the 2011 Fair Sentencing Act Guideline Amendment. Staff then compared that rate to the recidivism rate for a comparison group of offenders who would have been eligible to seek a reduced sentence under the 2011 amendment, but were released before the effective date of that amendment after serving their full prison terms less good time and other earned credits.

Key Findings

The Commission's report aims to answer the research question, "Did the reduced sentences for the FSA Retroactivity Group result in increased recidivism?".

Key findings of the Commission’s study are as follows:

  • The recidivism rates were virtually identical for offenders who were released early through retroactive application of the FSA Guideline Amendment and offenders who had served their full sentences before the FSA guideline reduction retroactively took effect. Over a three-year period following their release, the “FSA Retroactivity Group” and the “Comparison Group” each had a recidivism rate of 37.9 percent.

  • Among offenders who did recidivate, for both groups the category “court or supervision violation” was most often the most serious recidivist event reported. Approximately one-third of the offenders who recidivated in both groups (32.9% for the FSA Retroactivity Group and 30.8% for the Comparison Group) had court or supervision violation as their most serious recidivist event.

  • Among offenders who did recidivate, the time to recidivism for both groups were nearly identical. The median time to recidivism for offenders who recidivated in both groups was approximately 14½ months.

March 28, 2018 at 11:39 AM | Permalink


This is a study of rehabilitation, a form of quackery.

The study implies, keep the person longer, they will recidivate at the same rate, but later. Longer prison terms will delay recidivism, without influencing its rate.

Recidivism has no validity as an outcome measure. Only a small fraction of crime results in any arrest, prosecution, and conviction. The proper measurement is a survey of a good sample, with a guarantee of absolute immunity, and asking about the prior week. I commit so many crimes a day, I cannot recall all of them from yesterday. The same is true of any of the people here who have a job. For example, give me someone's laptop for an hour. I can send them to prison for decades and fine them $millions on the copyright infringements alone, never mind the embedded child porn they did not know about. If their jobs involve any government money, forget about it. I am not bringing up their tax returns. The sentences can stack to millennia.

Posted by: David Behar | Mar 28, 2018 11:53:41 AM

Behar, Behar, Behar, maybe next time you should read the report and methodology and actually ascertain the recidivism findings that are noted in the comprehensive report. Reading every other word or every other paragraph does you no good.

Posted by: atomicfrog | Mar 28, 2018 12:09:07 PM

Atomic. I have a policy of never reading this crap. Not even a word.

Posted by: David Behar | Mar 28, 2018 12:11:33 PM

Well, it's about time you posted some truth - LOL.....

Posted by: atomicfrog | Mar 28, 2018 12:12:58 PM

Atomic: As I said, worthless, invalid measurement. You just wasted my time. Thanks a lot. What are you a lawyer?

They missed thousands of crime committed by not using the valid method I suggested.

"For purposes of this study, recidivism is defined as any of the following criminal record events occurring within a three-year period following release from incarceration: • a re-conviction for a new offense; • a re-arrest with no case disposition information available; or • a violation or revocation of an offender’s supervised release. In other words, recidivism events include all reports of 1) arrests for new crimes that did not result in acquittal or dismissal of all charges, and 2) court or supervision violations that resulted in court action reported on criminal history records.15 All recidivism events within a three-year period after release were counted, including felonies, misdemeanors, and court and supervision violations, but minor traffic offenses were excluded."

Posted by: David Behar | Mar 28, 2018 12:15:10 PM

Once again, you are in la la land - from my stand point, the reports definition of recidivism is very clear, yours on the other hand, well, laughable and certainly impractical if not downright illogical. Sorry, you just don't get it and certainly don't care enough to get it.

Posted by: atomicfrog | Mar 28, 2018 12:23:47 PM

Atomic. Personal insults are not argument. You are not a lawyer. Go forth in peace. Have a blessed day.

Posted by: David Behar | Mar 28, 2018 1:08:28 PM

Post a comment

In the body of your email, please indicate if you are a professor, student, prosecutor, defense attorney, etc. so I can gain a sense of who is reading my blog. Thank you, DAB