« Senator Rand Paul criticizes new Sessions Memo while promising to reintroduce Justice Safety Valve Act | Main | Brennan Center releases "A Federal Agenda to Reduce Mass Incarceration" »

May 15, 2017

Reviewing some historical data on the federal prison population, total sentences imposed, and drug cases

Some of the copious commentary critical of the new Sessions Memo complains that he is "bringing back" the War on Drugs.  See, for example, Salon here, "Jeff Sessions is bringing back the drug war — and making it worse"; New York here, "Sessions Takes First Big Step Toward Bringing Back the War on Drugs."  I find this charge a bit curious because I do not think the drug war or its footprint on human lives ever really went away notwithstanding some recent efforts at the federal and state level to temper a bit its reach and impact. 

In an effort to try to see if the federal drug war at some point went away, and also driven by a desire to try to gauge the impact of federal charging policies before the Sessions Memo (as discussed here), I decided it might be useful to take a dive into US Sentencing Commission data over the past two decades to see what we could see.  The USSC has great yearly data assembled here going back to 1996, and basic federal prison population numbers are accessible here going back all the way to 1980.  Though my weak empirical skills and this imperfect blogging space will surely limit my ability to tell detailed data stories here effectively, I hope a few posts reviewing federal case processing and sentencing basics might be of some use and interest.  Here I will start with just the most basic of basics, historical data on the federal prison population, total sentences imposed, and drug cases:

Year        Federal Prison Population         Federal Sentences Imposed         Drug Sentences Imposed

1996                105,443                                        42,436                                    17,267

1998                122,316                                        50,754                                    20,368

2000                145,125                                        59,846                                    23,542

2002                163,436                                        64,366                                    25,920        

2004                179,895                                        70,068                                    24,532

2006                192,584                                        72,585                                    26,122

2008                201,668                                        76,478                                    25,500

2010                210,227                                        83,946                                    24,713

2012                218,687                                        84,173                                    25,712

2014                214,149                                        75,836                                    22,193

2016                192,170                                        67,742                                    19,945

May 15, 2017 at 07:07 PM | Permalink


I wonder about this emphasis on convictions. As the following link shows, immigration actually makes up 52% of all federal prosecutions.


In certain ways prosecutions are a more robust metric of federal priorities than convictions. The reason why is because from a defendant's POV a prosecution carries huge costs regardless of whether the prosecution results in their conviction or not. Indeed, in an important and often ignored way the prosecution is a sentence. It is a sentence in that it interferes with a person's liberty, their mental health, and their interpersonal relationships. In some cases (like sex crimes) being arrested alone is enough to destroy a reputation.

So in my view it is wrong to think of sentencing, and sentencing policy, as that which happens after a conviction. That's too formal for me.

Posted by: Daniel | May 15, 2017 7:41:14 PM

I largely agree with you, Daniel, that prosecutions are "a more robust metric" of federal priorities/actions than convictions and can be in a way a better mechanism for assessing the impact of charging policies. But especially if a driving concern for many activists is mass incarceration, convictions/sentences are a bit more tailored to that piece of the puzzle. And, practically speaking, TRAC hide a lot of its data behind a pay wall and thus its data is not as accessible as the US Sentencing Commission data.

That all said, you raise a good important point.

Posted by: Doug B | May 15, 2017 9:09:47 PM

"Though my weak empirical skills and this imperfect blogging space will surely limit my ability to tell detailed data stories here effectively, ..."

To keep it simple, you do not feel a difference smaller than a third at the gut level. No statistical calculation is needed. Differences can be highly statistically significant, but totally meaningless at the gut level. Worry about your gut feeling, and not any math.


That being said, the above is disgusting. The Computer Fraud and Abuse Act (CFAA) was enacted decades ago. These assholes are going after dealers who benefit society by spreading thousands of overdose deaths to criminals. Yet, they allow millions of hacking crimes. The next war will be fought on that battleground and not between $10 billion submarines.

I can't stand the stupidity of your profession. What will it take to wake you assholes up?

Posted by: David Behar | May 16, 2017 12:33:59 AM

I have to agree with David that overdose deaths are a benefit to society rather than cost.

Posted by: Soronel Haetir | May 16, 2017 1:07:03 AM

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